Expand model tag support: add GLM-5.1, simplify Anthropic IDs, scan tags anywhere in message

- Flink update_bars debouncing
- update_bars subscription idempotency bugfix
- Price decimal correction bugfix of previous commit
- Add GLM-5.1 model tag alongside renamed GLM-5
- Use short Anthropic model IDs (sonnet/haiku/opus) instead of full version strings
- Allow @tags anywhere in message content, not just at start
- Return hasOtherContent flag instead of trimmed rest string
- Only trigger greeting stream when tag has no other content
- Update workspace knowledge base references to platform/workspace and platform/shapes
- Hierarchical knowledge base catalog
- 151 Trading Strategies knowledge base articles
- Shapes knowledge base article
- MutateShapes tool instead of workspace patch

gateway/knowledge/trading/strategies/fixed-income/barbells.md

@@ -0,0 +1,50 @@
+---
+description: "A barbell portfolio holds bonds at two extreme maturities (short and long) to achieve a target duration while gaining higher convexity than an equivalent bullet, providing better protection against parallel yield curve shifts."
+tags: [fixed-income, duration, convexity, barbell, yield-curve]
+---
+
+# Barbells
+
+**Section**: 5.3 | **Asset Class**: Fixed Income | **Type**: Duration / Convexity
+
+## Overview
+A barbell concentrates holdings at two maturities: a short maturity T_1 and a long maturity T_2. It is a combination of two bullet strategies. For a given modified duration (matching a bullet at intermediate maturity T_*), the barbell achieves higher convexity, providing better protection against parallel yield shifts at the cost of lower overall yield.
+
+## Construction / Mechanics
+For a simple barbell of w_1 dollars in zero-coupon bonds with maturity T_1 and w_2 dollars with maturity T_2 (continuous compounding, constant yield Y), with price-adjusted weights w̃_1 = w_1·exp(-T_1·Y) and w̃_2 = w_2·exp(-T_2·Y):
+
+**Duration** (equals a bullet at T_*):
+```
+D = (w̃_1·T_1 + w̃_2·T_2) / (w̃_1 + w̃_2)                          (390)
+T_* = D_* = D                                                        (391)
+```
+
+**Convexity** (exceeds the equivalent bullet):
+```
+C = (w̃_1·T_1² + w̃_2·T_2²) / (w̃_1 + w̃_2)                        (392)
+C_* = T_*²                                                           (393)
+```
+
+The convexity advantage:
+```
+C - C_* = (w̃_1·w̃_2 / (w̃_1 + w̃_2)²) · (T_2 - T_1)² > 0          (394)
+```
+
+## Payoff / Return Profile
+- Higher convexity than an equivalent bullet means the barbell outperforms when yields move significantly in either direction (parallel shifts).
+- The long-maturity bonds benefit from high yields; the short-maturity bonds provide protection if rates rise (proceeds reinvested at higher rates).
+- Flattening of the yield curve (short-term rates rise relative to long-term) has a positive impact; steepening has a negative impact.
+
+## Key Parameters / Signals
+- T_1 (short maturity), T_2 (long maturity): the two maturities defining the barbell
+- w_1, w_2: dollar allocations to short and long maturities
+- Target duration D: matched to the equivalent bullet at T_*
+- Convexity advantage C - C_*: larger the spread T_2 - T_1, the greater the convexity benefit
+
+## Variations
+- Combine with duration matching to an intermediate bullet for controlled rate exposure.
+
+## Notes
+- Higher convexity comes at the expense of lower overall yield (yield curve typically slopes upward, so the mid-point bullet earns more carry).
+- Duration scales approximately linearly with maturity; convexity scales quadratically — this is why the barbell's convexity exceeds the equivalent bullet.
+- The barbell is more complex to manage than a bullet due to two distinct maturity exposures.

gateway/knowledge/trading/strategies/fixed-income/bond-immunization.md

@@ -0,0 +1,66 @@
+---
+description: "Bond immunization constructs a portfolio whose duration matches a future cash obligation's maturity, protecting the portfolio value against parallel yield curve shifts to meet a predetermined liability."
+tags: [fixed-income, duration, immunization, liability-matching, convexity]
+---
+
+# Bond Immunization
+
+**Section**: 5.5 | **Asset Class**: Fixed Income | **Type**: Duration / Liability Matching
+
+## Overview
+Bond immunization is used to ensure a portfolio can meet a predetermined future cash obligation F at time T_*. A portfolio is constructed so that its duration matches T_*, making its value insensitive to parallel shifts in the yield curve. It extends to matching convexity for additional protection with three bonds.
+
+## Construction / Mechanics
+
+**Total investment** P given a future obligation F at time T_*, constant yield Y, periodic compounding with period δ:
+```
+P = F / (1 + Yδ)^(T_*/δ)                                            (396)
+```
+
+**Two-bond immunization** (matches duration only):
+
+With two bonds of maturities T_1, T_2 and modified durations D_1, D_2, dollar allocations P_1, P_2:
+```
+P_1 + P_2 = P                                                        (397)
+P_1·D_1 + P_2·D_2 = P·D                                             (398)
+```
+where the target modified duration:
+```
+D = T_* / (1 + Yδ)                                                   (399)
+```
+
+**Three-bond immunization** (matches duration and convexity):
+
+With three bonds, durations D_1, D_2, D_3 and convexities C_1, C_2, C_3:
+```
+P_1 + P_2 + P_3 = P                                                  (400)
+P_1·D_1 + P_2·D_2 + P_3·D_3 = P·D                                   (401)
+P_1·C_1 + P_2·C_2 + P_3·C_3 = P·C                                   (402)
+```
+where the target convexity:
+```
+C = T_*(T_* + δ) / (1 + Yδ)²                                        (403)
+```
+
+## Payoff / Return Profile
+- Immunized portfolio is protected against parallel yield curve shifts: the gain/loss from price changes offsets the loss/gain from reinvestment rate changes.
+- Matching convexity (three-bond) provides additional protection against larger rate moves.
+- The portfolio value converges to F at time T_* under parallel shifts.
+
+## Key Parameters / Signals
+- T_*: maturity of the future cash obligation (target duration)
+- F: size of the future obligation
+- Y: assumed constant yield (all bonds assumed same yield — a simplification)
+- D, C: target modified duration and convexity
+
+## Variations
+- **Zero-coupon immunization**: purchase a single zero-coupon bond with maturity T_* — the simplest solution, but may not be available.
+- **Two-bond**: matches duration only; sufficient for small parallel shifts.
+- **Three-bond**: matches both duration and convexity; handles larger shifts.
+- Extension to non-parallel yield curve changes requires additional sophistication.
+
+## Notes
+- The assumption that all bonds have the same yield is a simplification; in practice yields differ across maturities and issuers.
+- The portfolio must be periodically rebalanced as the yield curve changes, incurring transaction costs.
+- Immunization protects against parallel shifts only; slope and curvature changes can still cause losses.
+- Non-parallel shifts, credit spread changes, and transaction costs all introduce complexity in practice.

gateway/knowledge/trading/strategies/fixed-income/bullets.md

@@ -0,0 +1,36 @@
+---
+description: "A bullet portfolio concentrates all bond holdings at a single target maturity, used to express a directional view on interest rates at a specific point on the yield curve."
+tags: [fixed-income, duration, bullet, yield-curve]
+---
+
+# Bullets
+
+**Section**: 5.2 | **Asset Class**: Fixed Income | **Type**: Duration / Directional
+
+## Overview
+In a bullet portfolio all bonds share the same maturity date T, targeting a specific segment of the yield curve. The strategy expresses a view on the direction of interest rates at that maturity. Bonds are typically purchased over time to mitigate timing risk from rate fluctuations.
+
+## Construction / Mechanics
+- Select a target maturity T based on the trader's interest rate outlook.
+- Purchase bonds of that maturity, potentially accumulating positions over time.
+- Hold to maturity or until the rate view is realized.
+
+Purchasing over time mitigates interest rate risk: if rates rise, later purchases capture higher yields; if rates fall, earlier purchases lock in higher yields.
+
+## Payoff / Return Profile
+- **Rates expected to fall** (bond prices rise): pick a longer maturity — longer bonds gain more in price from a given yield decline (higher duration).
+- **Rates expected to rise** (bond prices fall): pick a shorter maturity — shorter bonds lose less.
+- **Uncertain outlook**: diversify across maturities (barbell or ladder preferred).
+
+## Key Parameters / Signals
+- Target maturity T: the single maturity determining duration exposure
+- Modified duration: scales with T; determines price sensitivity to rate changes
+- Interest rate forecast: the primary signal driving maturity selection
+
+## Variations
+- Building the portfolio gradually over time to average in across different rate environments.
+
+## Notes
+- Concentrating at one maturity creates pure duration exposure with no convexity advantage.
+- Compared to a barbell with the same duration, a bullet has lower convexity, meaning it is more exposed to parallel yield curve shifts.
+- Suitable when the trader has a strong directional view on a specific maturity segment.

gateway/knowledge/trading/strategies/fixed-income/carry-factor.md

@@ -0,0 +1,56 @@
+---
+description: "The carry factor strategy buys bonds with the highest carry — the return earned as a bond rolls down the yield curve — combining bond yield income with the roll-down return from the yield curve's slope."
+tags: [fixed-income, factor, carry, roll-down, yield-curve]
+---
+
+# Carry Factor
+
+**Section**: 5.11 | **Asset Class**: Fixed Income | **Type**: Factor / Carry
+
+## Overview
+Carry in fixed income is the return from holding a bond as it "rolls down" the yield curve toward maturity. If the term structure is upward-sloping and stable, a bond's yield declines as its maturity shortens, causing a price appreciation on top of the coupon income. The carry factor strategy buys bonds in the top decile by carry and sells those in the bottom decile.
+
+## Construction / Mechanics
+
+**Carry** over horizon Δt for a bond with current maturity T:
+
+```
+C(t, t+Δt, T) = [P(t+Δt, T) - P(t, T)] / P(t, T)                  (413)
+```
+
+Under the assumption that the yield curve shape is constant (R(t,T) = f(T-t) only), the yield at t+Δt is R(t+Δt, T) = R(t, T-Δt), giving:
+
+```
+C(t, t+Δt, T) = R(t,T)·Δt + C_roll(t, t+Δt, T)                    (414)
+```
+
+Two components:
+1. **Yield income**: R(t,T)·Δt — the bond's current yield times the holding period
+2. **Roll-down return**:
+```
+C_roll(t, t+Δt, T) ≈ -ModD(t,T) · [R(t, T-Δt) - R(t, T)]          (415)
+```
+This is the price appreciation as the bond shortens in maturity by Δt along a static yield curve, estimated using modified duration.
+
+**Portfolio construction**: rank all bonds by C(t, t+Δt, T); long top decile, short bottom decile (zero-cost version).
+
+## Payoff / Return Profile
+- Earns yield income plus roll-down return when the yield curve is upward-sloping and stable.
+- Roll-down return is greatest in the steepest segments of the yield curve.
+- Underperforms or loses when the yield curve flattens, inverts, or shifts upward unexpectedly.
+
+## Key Parameters / Signals
+- R(t,T): current yield (income component)
+- ModD(t,T): modified duration (scales the roll-down component)
+- R(t, T-Δt) - R(t, T): slope of the yield curve at maturity T (steeper = more roll-down)
+- Δt: carry horizon (e.g., 1 month)
+
+## Variations
+- Long-only: buy top decile by carry (no short sales required).
+- Cross-asset carry: extend the same framework to other fixed income markets (government bonds, credit, etc.).
+
+## Notes
+- The static yield curve assumption simplifies computation; actual carry will differ if the curve shifts.
+- For financed portfolios, R(t,T) is replaced by R(t,T) - r_f (excess yield over the risk-free rate) in the income component, but this does not affect portfolio weights.
+- High-carry bonds tend to have longer maturities in an upward-sloping curve environment, so the carry factor has implicit duration exposure.
+- Carry and roll-down are sometimes separated as distinct signals; roll-down alone favors bonds in the steepest curve segments regardless of yield level.

gateway/knowledge/trading/strategies/fixed-income/cds-basis-arbitrage.md

@@ -0,0 +1,55 @@
+---
+description: "CDS basis arbitrage exploits the mispricing between a bond's credit spread and its CDS spread — when the CDS basis is negative (bond spread too high), buy the bond and buy CDS protection to lock in a risk-free profit."
+tags: [fixed-income, arbitrage, cds, credit-spread, basis]
+---
+
+# CDS Basis Arbitrage
+
+**Section**: 5.14 | **Asset Class**: Fixed Income | **Type**: Arbitrage / Credit
+
+## Overview
+A credit default swap (CDS) provides insurance against default on a bond. In theory, the CDS spread should equal the bond yield spread over the risk-free rate, making the insured bond equivalent to a risk-free instrument. The CDS basis is the difference between these two spreads, and deviations from zero create arbitrage opportunities.
+
+## Construction / Mechanics
+
+**CDS basis**:
+```
+CDS basis = CDS spread - bond spread                                 (417)
+```
+
+where bond spread = bond yield - risk-free rate.
+
+**Arbitrage logic**:
+- CDS spread should ≈ bond spread (both represent compensation for default risk)
+- If CDS basis ≠ 0 (and |basis| exceeds transaction costs), an arbitrage opportunity exists
+
+**Negative basis trade** (most common):
+- CDS basis < 0: bond spread > CDS spread → bond is relatively cheap
+- Trade: **buy the bond** (receive the high spread) + **buy CDS protection** (pay the lower CDS spread)
+- Net P&L per dollar of insured debt: bond spread - CDS spread = -basis > 0
+- Result: a nearly risk-free positive carry, since the CDS makes the bond effectively risk-free
+
+**Positive basis trade** (less common in practice):
+- CDS basis > 0: CDS spread > bond spread → CDS protection is expensive relative to bond
+- Trade: sell the bond + sell CDS protection (write CDS)
+- In practice, this often means unwinding an existing position (already owning both the bond and CDS)
+
+## Payoff / Return Profile
+- Earns the absolute value of the CDS basis as a near-riskless spread.
+- Position closed when basis converges back to zero.
+- The trade is essentially a carry trade: positive carry from the basis for as long as it persists.
+
+## Key Parameters / Signals
+- CDS basis = CDS spread - bond spread: the primary signal
+- Transaction cost threshold: |basis| must exceed bid-ask spreads and financing costs
+- Sign of basis: negative → buy bond + buy CDS; positive → sell bond + sell CDS
+
+## Variations
+- **Synthetic bond replication**: CDS + risk-free bond (e.g., Treasury repo) replicates a corporate bond; mispricing between the two creates the arbitrage.
+
+## Notes
+- CDS protection makes the bond synthetically risk-free, but counterparty risk on the CDS remains.
+- Negative basis arbitrage requires financing the bond purchase (repo market); the repo rate affects net P&L.
+- The CDS basis can persist or widen during stress periods (e.g., 2008 financial crisis) before eventually converging, creating significant mark-to-market losses in the interim.
+- Liquidity risk: corporate bonds may be illiquid, making it difficult to close the position at fair value.
+- In the positive basis case, selling a corporate bond short is operationally challenging.

gateway/knowledge/trading/strategies/fixed-income/dollar-duration-neutral-butterfly.md

@@ -0,0 +1,44 @@
+---
+description: "A dollar-duration-neutral butterfly combines a long barbell (short T_1 and long T_3 maturities) with a short bullet (intermediate T_2) at zero net cost, immunizing against parallel yield curve shifts to profit from yield curve curvature changes."
+tags: [fixed-income, butterfly, duration-neutral, yield-curve, curvature]
+---
+
+# Dollar-Duration-Neutral Butterfly
+
+**Section**: 5.6 | **Asset Class**: Fixed Income | **Type**: Yield Curve / Curvature
+
+## Overview
+The dollar-duration-neutral butterfly is a zero-cost combination of a long barbell (long T_1 and T_3 maturity bonds) and a short bullet (short the T_2 intermediate maturity bond), where T_1 < T_2 < T_3. Both zero cost (dollar neutrality) and dollar-duration neutrality conditions are imposed, immunizing the portfolio against parallel yield curve shifts. The strategy profits from changes in yield curve curvature.
+
+## Construction / Mechanics
+Let P_1, P_2, P_3 be the dollar amounts invested in the three bonds, and D_1, D_2, D_3 their modified durations.
+
+**Zero-cost** (dollar neutrality): the long barbell finances the short bullet position:
+```
+P_1 + P_3 = P_2                                                      (404)
+```
+
+**Dollar-duration neutrality** (parallel shift immunity):
+```
+P_1·D_1 + P_3·D_3 = P_2·D_2                                         (405)
+```
+
+These two equations determine P_1 and P_3 given P_2.
+
+## Payoff / Return Profile
+- Profits when the yield curve becomes more curved (humped): the intermediate yield rises relative to the wings, or the wings fall relative to the body.
+- Immune to small parallel shifts in the yield curve (both level and dollar-duration matched).
+- Exposed to changes in the slope and curvature of the yield curve.
+
+## Key Parameters / Signals
+- T_1 (short wing), T_2 (body), T_3 (long wing): the three maturities; T_1 < T_2 < T_3
+- D_1, D_2, D_3: modified durations of the three bonds
+- P_2: the reference position size (determines P_1 and P_3 via the two constraints)
+
+## Variations
+- See also: fifty-fifty butterfly (5.7) and regression-weighted butterfly (5.8), which relax the zero-cost condition.
+
+## Notes
+- Dollar-duration neutrality (Eq. 405) protects against parallel shifts only; non-parallel changes in slope or curvature can still generate losses or gains.
+- The zero-cost constraint (Eq. 404) means no initial capital is required, making it attractive as an overlay strategy.
+- In practice, bid-ask spreads, financing costs, and liquidity differences across maturities affect profitability.

gateway/knowledge/trading/strategies/fixed-income/fifty-fifty-butterfly.md

@@ -0,0 +1,46 @@
+---
+description: "A fifty-fifty butterfly sets equal dollar durations on both wings of the barbell, making it approximately neutral to small yield curve steepening and flattening while remaining dollar-duration neutral, trading zero-cost for curve-neutrality."
+tags: [fixed-income, butterfly, duration-neutral, yield-curve, curvature]
+---
+
+# Fifty-Fifty Butterfly
+
+**Section**: 5.7 | **Asset Class**: Fixed Income | **Type**: Yield Curve / Curvature
+
+## Overview
+The fifty-fifty butterfly is a variation of the dollar-duration-neutral butterfly that equalizes the dollar durations of the two wings (short-maturity and long-maturity positions). This makes the strategy approximately neutral to small steepening and flattening of the yield curve (not just parallel shifts), at the cost of no longer being dollar-neutral (it is not zero-cost). It is also known as the "neutral curve butterfly."
+
+## Construction / Mechanics
+Using the same notation as the dollar-duration-neutral butterfly (Section 5.6), with modified durations D_1, D_2, D_3 and dollar positions P_1, P_2, P_3:
+
+**Equal wing dollar durations**:
+```
+P_1·D_1 = P_3·D_3 = (1/2)·P_2·D_2                                  (406)
+```
+
+This implies dollar-duration neutrality is preserved:
+```
+P_1·D_1 + P_3·D_3 = P_2·D_2
+```
+
+But the zero-cost condition P_1 + P_3 = P_2 is generally not satisfied.
+
+## Payoff / Return Profile
+- Approximately neutral to small steepening and flattening of the yield curve: the spread change between the body (T_2) and the short wing (T_1) equals the spread change between the body and the long wing (T_3).
+- Still immune to parallel shifts (dollar-duration neutral).
+- Profits from curvature changes: if the body cheapens relative to both wings, the position gains.
+
+## Key Parameters / Signals
+- P_1·D_1 = P_3·D_3 = (1/2)·P_2·D_2: the defining equal-wing constraint
+- T_1 < T_2 < T_3: the three maturities
+- Net cost P_2 - P_1 - P_3: non-zero unlike the dollar-duration-neutral butterfly
+
+## Variations
+- Dollar-duration-neutral butterfly (Section 5.6): zero-cost but not curve-neutral.
+- Regression-weighted butterfly (Section 5.8): uses empirical β to account for differential yield volatility across the curve.
+
+## Notes
+- The name "fifty-fifty" refers to the equal split of the body's dollar duration between the two wings.
+- Curve-neutrality is approximate and holds only for small parallel steepening/flattening moves.
+- The non-zero cost means the trader must finance the net position, which has carry implications.
+- Short-term rates are empirically more volatile than long-term rates, which limits the curve-neutrality assumption; this motivates the regression-weighted butterfly.

gateway/knowledge/trading/strategies/fixed-income/generalities.md

@@ -0,0 +1,100 @@
+---
+description: "Background concepts for fixed income instruments: zero-coupon bonds, coupon bonds, floating rate bonds, swaps, duration, and convexity — the foundational mechanics underlying all fixed income strategies."
+tags: [fixed-income, background, duration, convexity, swaps]
+---
+
+# Fixed Income Generalities
+
+**Section**: 5.1 | **Asset Class**: Fixed Income | **Type**: Background / Reference
+
+## Overview
+Fixed income instruments are promises to pay cash flows at future dates, priced today as the present value of those flows. The yield of a bond summarizes its return as a single annualized rate. Duration and convexity characterize how bond prices respond to interest rate changes, and are the primary risk-management tools for fixed income portfolios.
+
+## Construction / Mechanics
+
+### 5.1.1 Zero-Coupon Bonds
+A zero-coupon (discount) bond with maturity T pays $1 at time T. Its price at time t is P(t,T), with P(T,T) = 1. The continuously compounded yield is:
+
+```
+R(t,T) = -ln(P(t,T)) / (T - t)                                     (374)
+```
+
+### 5.1.2 Coupon Bonds
+A coupon bond pays principal $1 at maturity T plus n coupon payments of amount kδ at times T_i = T_0 + iδ (i = 1,...,n), where δ is the payment period. Price at time t:
+
+```
+P_c(t,T) = P(t,T) + kδ Σ_{i=I(t)}^n P(t,T_i)                      (375)
+```
+
+where I(t) = min(i : t < T_i). At issuance (t = T_0), the par coupon rate is:
+
+```
+k = (1 - P(T_0,T)) / (δ Σ_{i=1}^n P(T_0,T_i))                     (377)
+```
+
+### 5.1.3 Floating Rate Bonds
+Coupon payments are based on LIBOR. The LIBOR rate at T_{i-1} for period [T_{i-1}, T_i] is:
+
+```
+L(T_{i-1}) = (1/δ) [1/P(T_{i-1},T_i) - 1]                         (378)
+```
+
+The coupon paid at T_i is X_i = L(T_{i-1})δ = 1/P(T_{i-1},T_i) - 1. The total value at T_0:
+
+```
+V_0 = 1 - [P(T_0,T_n) - P(T_0,T)]                                  (380)
+```
+
+If T = T_n then V_0 = 1 (the bond prices at par).
+
+### 5.1.4 Swaps
+An interest rate swap exchanges fixed rate payments for floating (LIBOR) payments. A long swap = long fixed coupon bond + short floating rate bond. The fixed rate giving zero initial value:
+
+```
+k = (1 - P(T_0,T_n)) / (δ Σ_{i=1}^n P(T_0,T_i))                   (383)
+```
+
+### 5.1.5 Duration and Convexity
+**Macaulay duration** is the present-value-weighted average maturity of cash flows:
+
+```
+MacD(t,T) = (1/P_c(t,T)) [(T-t)P(t,T) + kδ Σ_{i=I(t)}^n (T_i-t)P(t,T_i)]   (384)
+```
+
+**Modified duration** measures relative price sensitivity to parallel yield shifts:
+
+```
+ModD(t,T) = -∂ln(P_c(t,T)) / ∂R(t,T)                               (385)
+```
+
+For constant yield Y with periodic compounding: ModD = MacD / (1 + Yδ).  
+Approximate price change: ΔP_c/P_c ≈ -ModD · ΔR
+
+**Dollar duration** measures absolute price sensitivity:
+
+```
+DD(t,T) = -∂P_c(t,T)/∂R(t,T) = ModD(t,T) · P_c(t,T)               (387)
+```
+
+**Convexity** captures nonlinear (second-order) effects:
+
+```
+C(t,T) = -(1/P_c(t,T)) · ∂²P_c(t,T)/∂R(t,T)²                      (388)
+```
+
+Full second-order approximation:
+
+```
+ΔP_c/P_c ≈ -ModD·ΔR + (1/2)·C·(ΔR)²                               (389)
+```
+
+## Key Parameters / Signals
+- **Yield R(t,T)**: inverse of price; drives all valuation
+- **Modified duration**: primary interest rate risk metric; scales approximately linearly with maturity
+- **Dollar duration**: used for hedging and portfolio construction
+- **Convexity**: scales approximately quadratically with maturity; higher convexity = better protection against parallel yield shifts at the cost of lower yield
+
+## Notes
+- Duration and convexity formulas assume parallel shifts in the yield curve; non-parallel shifts require more sophisticated treatment.
+- Floating rate bonds priced at par (V_0 = 1) when T = T_n because the variable coupons replicate rolling T-bond investments.
+- Periodic vs. continuous compounding: MacD and ModD coincide under continuous compounding; differ under periodic compounding by factor (1 + Yδ).

gateway/knowledge/trading/strategies/fixed-income/ladders.md

@@ -0,0 +1,43 @@
+---
+description: "A ladder portfolio holds bonds spread evenly across n equidistant maturities to diversify interest rate and reinvestment risk while maintaining an approximately constant duration through systematic roll-down."
+tags: [fixed-income, duration, ladder, diversification, yield-curve]
+---
+
+# Ladders
+
+**Section**: 5.4 | **Asset Class**: Fixed Income | **Type**: Duration-Targeting / Diversification
+
+## Overview
+A ladder holds bonds with (roughly) equal capital allocations across n different maturities T_i (i = 1,...,n), where maturities are equidistant: T_{i+1} = T_i + δ. The strategy maintains an approximately constant duration by selling shorter-maturity bonds as they near maturity and replacing them with new longer-maturity bonds. It diversifies both interest rate risk and reinvestment risk.
+
+## Construction / Mechanics
+- Allocate roughly equal capital to each rung T_i, i = 1,...,n (n is sizable, e.g., n = 10).
+- Equidistant maturities: T_{i+1} = T_i + δ.
+- Average (effective) maturity of the portfolio:
+
+```
+T = (1/n) Σ_{i=1}^n T_i                                             (395)
+```
+
+- As the shortest rung approaches maturity, sell it and purchase a new bond at the longest maturity, maintaining the ladder structure.
+- Also generates regular income from coupon payments across all rungs.
+
+## Payoff / Return Profile
+- Higher average maturity T → higher income (upward-sloping yield curve), but also higher interest rate risk.
+- Rolling shorter bonds into longer bonds continuously captures roll-down return.
+- Diversification across maturities smooths the impact of rate moves: if rates rise, maturing short bonds are reinvested at higher rates; if rates fall, longer bonds appreciate.
+
+## Key Parameters / Signals
+- n: number of rungs (more rungs = more diversification)
+- δ: spacing between maturities
+- T (average maturity): determines the income/risk trade-off
+- Equal capital allocation per rung: ensures no concentration at any maturity
+
+## Variations
+- Unequal allocations tilting toward shorter or longer maturities (incorporating a partial bullet or barbell bias).
+
+## Notes
+- The ladder avoids the concentration risk of bullets and barbells, making it suitable for investors uncertain about the rate environment.
+- The constant-duration property is approximate; exact duration changes as bonds age and are replaced.
+- Reinvestment risk is diversified: proceeds from maturing bonds are spread across the yield curve over time rather than all reinvested at once.
+- Transaction costs from regular rolling must be weighed against the diversification and roll-down benefits.

gateway/knowledge/trading/strategies/fixed-income/low-risk-factor.md

@@ -0,0 +1,44 @@
+---
+description: "The low-risk factor strategy buys bonds with lower risk (shorter maturity and higher credit rating) within a credit tier, exploiting the empirical anomaly that lower-risk bonds outperform higher-risk bonds on a risk-adjusted basis."
+tags: [fixed-income, factor, low-risk, credit, anomaly]
+---
+
+# Low-Risk Factor
+
+**Section**: 5.9 | **Asset Class**: Fixed Income | **Type**: Factor / Anomaly
+
+## Overview
+Empirical evidence suggests that lower-risk bonds tend to outperform higher-risk bonds on a risk-adjusted basis (the "low-risk anomaly"), mirroring a similar effect in equities. "Riskiness" in fixed income is measured by credit rating and maturity. The strategy builds long portfolios of the lowest-risk bonds within a given credit tier.
+
+## Construction / Mechanics
+Portfolio construction uses two risk dimensions:
+
+1. **Credit rating**: separates the investment universe into quality tiers.
+   - Investment Grade (IG): credit ratings AAA through A-.
+   - High Yield (HY): credit ratings BB+ through B-.
+
+2. **Maturity (duration)**: within each credit tier, rank bonds by maturity and take the **bottom decile** (shortest maturities = lowest duration risk).
+
+Example portfolios:
+- IG low-risk: Investment Grade bonds (AAA–A-), bottom decile by maturity.
+- HY low-risk: High Yield bonds (BB+–B-), bottom decile by maturity.
+
+## Payoff / Return Profile
+- Earns a risk-adjusted premium by being long the lowest-risk bonds in each tier.
+- Outperforms the broad credit market on a Sharpe ratio basis due to the low-risk anomaly.
+- Returns are driven by credit spread compression and coupon income, with lower sensitivity to interest rate moves (short maturity).
+
+## Key Parameters / Signals
+- Credit rating tier: AAA–A- (IG) or BB+–B- (HY)
+- Maturity rank: bottom decile selects shortest-maturity bonds
+- Risk-adjusted return (Sharpe ratio): primary evaluation metric
+
+## Variations
+- Can be combined with a short position in the top-risk decile (highest maturity within the tier) to create a long-short low-risk factor.
+- Risk metrics beyond credit rating and maturity (e.g., option-adjusted spread, liquidity) can be incorporated.
+
+## Notes
+- The low-risk anomaly in bonds mirrors the similar effect documented in equities but is driven by different mechanisms (credit and duration rather than beta).
+- Separating IG and HY tiers is important because the risk-return relationship differs significantly between investment grade and speculative grade.
+- Liquidity may be lower for short-maturity high-yield bonds, increasing transaction costs.
+- The strategy is typically implemented as a long-only portfolio; short positions in corporate bonds are operationally difficult.

gateway/knowledge/trading/strategies/fixed-income/regression-weighted-butterfly.md

@@ -0,0 +1,53 @@
+---
+description: "A regression-weighted butterfly uses an empirically estimated β to account for the higher volatility of short-term rates relative to long-term rates, improving yield-curve-neutrality beyond the fifty-fifty butterfly."
+tags: [fixed-income, butterfly, duration-neutral, yield-curve, curvature, regression]
+---
+
+# Regression-Weighted Butterfly
+
+**Section**: 5.8 | **Asset Class**: Fixed Income | **Type**: Yield Curve / Curvature
+
+## Overview
+Empirically, short-term interest rates are significantly more volatile than long-term rates. The regression-weighted butterfly accounts for this by weighting the short wing's dollar duration by a factor β > 1, estimated from historical data via regression. This produces better curve-neutrality than the fifty-fifty butterfly in practice.
+
+## Construction / Mechanics
+Using positions P_1, P_2, P_3 with modified durations D_1, D_2, D_3 (T_1 < T_2 < T_3):
+
+**Dollar-duration neutrality** (parallel shift immunity):
+```
+P_1·D_1 + P_3·D_3 = P_2·D_2                                         (407)
+```
+
+**Regression-weighted curve-neutrality**:
+```
+P_1·D_1 = β · P_3·D_3                                               (408)
+```
+
+where β > 1 is the regression coefficient from regressing the spread change between the body (T_2) and the short wing (T_1) on the spread change between the body and the long wing (T_3), using historical data.
+
+## Payoff / Return Profile
+- Immune to both parallel shifts (407) and, approximately, to yield curve slope changes in proportion β.
+- Profits from yield curve curvature moves: gains when the body yields rise relative to the wings.
+- More robust curve-neutrality than the fifty-fifty butterfly in practice due to the empirically calibrated β.
+
+## Key Parameters / Signals
+- β: regression coefficient (typically β > 1, calibrated from historical spread data)
+- P_1, P_3: determined by solving (407) and (408) given P_2
+- T_1, T_2, T_3: the three maturity points on the yield curve
+
+## Variations
+
+### 5.8.1 Maturity-Weighted Butterfly
+Instead of estimating β from historical regressions, it is set analytically from the three bond maturities:
+
+```
+β = (T_2 - T_1) / (T_3 - T_2)                                       (409)
+```
+
+This is proportional to the ratio of the short-wing maturity distance to the long-wing maturity distance from the body. It is a simpler, model-based alternative that does not require historical calibration.
+
+## Notes
+- β is empirically greater than 1 because short-term rates fluctuate more than long-term rates; the short wing therefore needs less dollar duration to hedge the same spread move.
+- The regression β should be re-estimated periodically as the volatility relationship between short and long rates can change over time.
+- The maturity-weighted variant (5.8.1) provides a model-based β that requires no estimation but may not capture the true empirical volatility asymmetry.
+- All butterfly strategies share the exposure to transaction costs, financing costs, and bid-ask spreads that can erode theoretical curve-neutrality profits.

gateway/knowledge/trading/strategies/fixed-income/rolling-down-yield-curve.md

@@ -0,0 +1,52 @@
+---
+description: "Rolling down the yield curve buys long- or medium-term bonds in the steepest segment of the yield curve and holds them while they appreciate as they shorten in maturity, then reinvests proceeds into new steepest-segment bonds."
+tags: [fixed-income, roll-down, yield-curve, carry, duration]
+---
+
+# Rolling Down the Yield Curve
+
+**Section**: 5.12 | **Asset Class**: Fixed Income | **Type**: Carry / Roll-Down
+
+## Overview
+The rolling down the yield curve strategy captures the roll-down component C_roll of bond returns by purchasing bonds in the steepest segments of the yield curve and holding them while their maturity shortens, causing price appreciation. Bonds are sold before maturity and the proceeds reinvested in new long/medium-term bonds from the same steep segment.
+
+## Construction / Mechanics
+The roll-down return over horizon Δt:
+
+```
+C_roll(t, t+Δt, T) ≈ -ModD(t,T) · [R(t, T-Δt) - R(t, T)]          (415)
+```
+
+This is maximized when:
+- **ModD(t,T)** is large (longer-maturity bonds have higher duration)
+- **R(t, T-Δt) - R(t, T) < 0** (yield declines as maturity shortens — upward-sloping curve)
+- The magnitude |R(t, T-Δt) - R(t, T)| is large (steep segment of the curve)
+
+**Strategy mechanics**:
+1. Identify the steepest segment(s) of the yield curve.
+2. Buy long- or medium-term bonds from those segments.
+3. Hold while they "roll down" the curve (their maturity shrinks and yield declines).
+4. Sell before maturity approaches (before they enter a flatter/shorter segment).
+5. Reinvest proceeds into new long/medium-term bonds from the steep segment.
+
+## Payoff / Return Profile
+- Earns roll-down return C_roll in addition to yield income R(t,T)·Δt.
+- Total carry C(t, t+Δt, T) = R(t,T)·Δt + C_roll(t, t+Δt, T).
+- Profits maximized in steeply upward-sloping yield curves.
+- Loses money when the yield curve flattens, inverts, or when long-end yields rise (parallel upward shift).
+
+## Key Parameters / Signals
+- Yield curve slope: identifies which segments offer the most roll-down return
+- Modified duration: amplifies the roll-down return
+- Holding horizon Δt: determines how far down the curve the bond rolls before sale
+- Curve stability: strategy depends on curve shape remaining approximately stable
+
+## Variations
+- Pure roll-down: focus exclusively on C_roll, ignoring yield income (selects steepest curve segments regardless of absolute yield level).
+- Combined carry + roll: as in the carry factor strategy (5.11), which uses total C as the signal.
+
+## Notes
+- The yield curve must be upward-sloping for roll-down to be positive; in a flat or inverted curve the roll-down may be zero or negative.
+- Transaction costs from repeated roll-overs must be weighed against the roll-down income.
+- The strategy has implicit duration risk: long/medium bonds lose value in a rising rate environment, which can more than offset the roll-down gain.
+- Steepest curve segments often occur at the short to medium end (e.g., 2-10 year part of the Treasury curve) and can shift over time with monetary policy.

gateway/knowledge/trading/strategies/fixed-income/swap-spread-arbitrage.md

@@ -0,0 +1,59 @@
+---
+description: "Swap spread arbitrage takes a long (short) position in an interest rate swap versus a short (long) position in a Treasury bond of the same maturity, profiting from the difference between the swap rate, Treasury yield, LIBOR, and the repo rate."
+tags: [fixed-income, arbitrage, swap, libor, treasury, spread]
+---
+
+# Swap Spread Arbitrage
+
+**Section**: 5.15 | **Asset Class**: Fixed Income | **Type**: Arbitrage / Rates
+
+## Overview
+The swap spread arbitrage is a dollar-neutral strategy that combines a long (or short) position in an interest rate swap with an offsetting short (or long) position in a Treasury bond of the same maturity. It profits from the spread between the swap fixed rate and the Treasury yield, net of financing costs (LIBOR vs. repo rate). The strategy is essentially a bet on the direction of LIBOR relative to the repo rate.
+
+## Construction / Mechanics
+
+**Instruments**:
+- Interest rate swap: receive fixed rate r_swap, pay floating LIBOR L(t)
+- Treasury bond: short the bond (financed at repo rate r(t))
+
+**Per-dollar-invested P&L rate**:
+```
+C(t) = ±[C_1 - C_2(t)]                                             (418)
+C_1  = r_swap - Y_Treasury                                           (419)
+C_2(t) = L(t) - r(t)                                                (420)
+```
+
+where:
+- C_1: constant spread = swap fixed rate minus Treasury yield (the swap spread)
+- C_2(t): floating spread = LIBOR minus repo rate
+- Plus sign: long swap strategy (receive fixed, short Treasury)
+- Minus sign: short swap strategy (pay fixed, long Treasury)
+
+**Long swap strategy** (plus sign):
+- Receive r_swap (fixed leg of swap) + short Treasury (financed at repo) → pay Y_Treasury + repo rate
+- Profitable if C_2(t) = L(t) - r(t) < C_1
+
+**Short swap strategy** (minus sign):
+- Pay r_swap (fixed leg) + long Treasury (funded at repo) → receive Y_Treasury
+- Profitable if C_2(t) = L(t) - r(t) > C_1
+
+## Payoff / Return Profile
+- The long swap strategy profits if LIBOR falls (C_2 decreases below C_1).
+- The short swap strategy profits if LIBOR rises (C_2 increases above C_1).
+- This is fundamentally a **LIBOR bet**: the trade profits or loses based on the LIBOR-repo spread relative to the constant swap spread C_1.
+
+## Key Parameters / Signals
+- C_1 = r_swap - Y_Treasury: the swap spread (constant at trade inception)
+- C_2(t) = L(t) - r(t): the LIBOR-repo spread (time-varying)
+- Net P&L driver: ±(C_1 - C_2(t)); direction depends on long vs. short swap position
+
+## Variations
+- Adjust maturity of the swap and Treasury bond to target different parts of the yield curve.
+- Pair with CDS basis trades for multi-leg credit/rates arbitrage.
+
+## Notes
+- The strategy is dollar-neutral (the swap and Treasury position offset each other in notional terms).
+- LIBOR risk is the dominant risk: unexpected changes in LIBOR (e.g., central bank policy shifts, bank credit stress) directly affect P&L.
+- The repo rate r(t) can vary and introduces additional uncertainty in C_2(t).
+- With the transition away from LIBOR to SOFR and other risk-free rates, the mechanics of this strategy are evolving.
+- Counterparty risk on the swap and margin requirements must be accounted for in practice.

gateway/knowledge/trading/strategies/fixed-income/value-factor.md

@@ -0,0 +1,66 @@
+---
+description: "The value factor strategy for bonds selects bonds with the highest actual credit spread relative to a theoretically predicted spread from a cross-sectional regression, going long undervalued bonds in the top decile."
+tags: [fixed-income, factor, value, credit-spread, regression]
+---
+
+# Value Factor
+
+**Section**: 5.10 | **Asset Class**: Fixed Income | **Type**: Factor / Value
+
+## Overview
+"Value" in fixed income is defined by comparing a bond's observed credit spread to a theoretically predicted (fair value) credit spread. Bonds trading with a spread significantly above their predicted fair value are cheap (high value); those below are expensive. The strategy buys the top-decile bonds by value score.
+
+## Construction / Mechanics
+
+**Step 1: Estimate fair value spreads** via a cross-sectional linear regression across N bonds (i = 1,...,N):
+
+```
+S_i = Σ_{r=1}^K β_r · I_{ir} + γ · T_i + ε_i                      (410)
+```
+
+where:
+- S_i: observed credit spread of bond i (bond yield minus risk-free rate)
+- I_{ir}: dummy variable = 1 if bond i has credit rating r, 0 otherwise (K ≤ 21 ratings)
+- T_i: maturity of bond i
+- β_r, γ: regression coefficients (note: no separate intercept since Σ_r I_{ir} = 1 for each bond)
+- ε_i: regression residual
+
+The constraint:
+```
+Σ_{r=1}^K I_{ir} = 1   for all i                                    (412)
+```
+(each bond has exactly one credit rating, so the intercept is absorbed into the rating dummies)
+
+**Step 2: Compute fair value spread**:
+```
+S_i* = S_i - ε_i                                                     (411)
+```
+(the fitted value from the regression)
+
+**Step 3: Compute value score** — either:
+- V_i = ln(S_i / S_i*), or
+- V_i = ε_i / S_i* = S_i / S_i* - 1
+
+**Step 4: Select portfolio** — long bonds in the top decile by V_i (most undervalued).
+
+## Payoff / Return Profile
+- Profits when cheap bonds (high V_i) revert toward fair value, compressing their spreads.
+- Returns driven by credit spread compression and coupon income.
+- The strategy assumes mean-reversion in credit spreads around their rating- and maturity-implied fair value.
+
+## Key Parameters / Signals
+- S_i: observed credit spread (bond yield minus risk-free rate)
+- S_i*: fair value credit spread from cross-sectional regression
+- V_i = ln(S_i/S_i*) or V_i = S_i/S_i* - 1: value score
+- Top decile by V_i: the portfolio selection criterion
+
+## Variations
+- Long-short: long top decile (cheap bonds), short bottom decile (expensive bonds).
+- Separate regressions for Investment Grade and High Yield universes.
+- Additional cross-sectional controls (e.g., industry, liquidity) can be added as regressors.
+
+## Notes
+- "Value" in fixed income is harder to define than in equities because bonds have finite lifetimes and their spreads are heavily influenced by credit ratings and maturity.
+- The cross-sectional regression should be run on bonds within a comparable universe (e.g., only IG or only HY) to ensure meaningful comparisons.
+- Credit spread data may be noisy; outliers from bonds near distress can distort the regression.
+- Shorting corporate bonds is operationally challenging; the strategy is often implemented long-only.

gateway/knowledge/trading/strategies/fixed-income/yield-curve-spread.md

@@ -0,0 +1,56 @@
+---
+description: "Yield curve spread strategies (flatteners and steepeners) trade the difference in yields between two maturities of the same issuer, going short the spread when rates are expected to rise and long when rates are expected to fall."
+tags: [fixed-income, yield-curve, spread, flattener, steepener, duration]
+---
+
+# Yield Curve Spread (Flatteners & Steepeners)
+
+**Section**: 5.13 | **Asset Class**: Fixed Income | **Type**: Yield Curve / Spread
+
+## Overview
+Yield curve spread strategies trade the yield spread between two bonds of the same issuer at different maturities. If interest rates are expected to rise, the yield curve is expected to flatten (short end rises more than long end); if rates are expected to fall, the curve steepens. The strategy goes short the spread (flattener) or long the spread (steepener) accordingly.
+
+## Construction / Mechanics
+
+**Yield curve spread**: the difference in yields between a longer-maturity bond (back leg) and a shorter-maturity bond (front leg) of the same issuer:
+```
+Spread = Y(back leg, long maturity) - Y(front leg, short maturity)
+```
+
+**Trading rule**:
+```
+Rule = { Flattener: Short spread   if interest rates expected to rise
+        { Steepener: Buy spread    if interest rates expected to fall      (416)
+```
+
+**Position construction**:
+- **Short the spread (flattener)**: sell shorter-maturity bonds (front leg) + buy longer-maturity bonds (back leg).
+- **Buy the spread (steepener)**: buy shorter-maturity bonds (front leg) + sell longer-maturity bonds (back leg).
+
+**Dollar-duration matching**: to immunize against small parallel shifts, match the dollar durations of the front and back legs:
+```
+P_front · D_front = P_back · D_back
+```
+
+Without duration matching, a parallel shift in the yield curve can generate significant losses.
+
+## Payoff / Return Profile
+- **Flattener profits** when the curve flattens: short-end yields rise more than long-end yields (or long-end falls more than short-end).
+- **Steepener profits** when the curve steepens: long-end yields rise more than short-end (or short-end falls more than long-end).
+- Dollar-duration-neutral construction limits losses from parallel yield curve moves.
+
+## Key Parameters / Signals
+- Yield curve slope: R(long maturity) - R(short maturity) — the key signal
+- Front leg maturity T_1, back leg maturity T_2: define the segment being traded
+- Modified durations D_1, D_2: used for dollar-duration matching
+- Interest rate outlook: the primary driver of direction (flattener vs. steepener)
+
+## Variations
+- **Curve trades across issuers**: trading the slope difference between two issuers (adds credit spread risk).
+- **Butterfly trades**: extend the two-leg spread to a three-leg position to trade curvature rather than slope (see Sections 5.6–5.8).
+
+## Notes
+- Parallel yield curve shifts can cause losses if dollar durations are not matched; duration-matching is essential for a pure slope bet.
+- Even with duration matching, large parallel moves (exceeding the immunization approximation) can generate losses due to convexity differences between legs.
+- The strategy is exposed to idiosyncratic supply-and-demand effects at specific maturities (e.g., Treasury auction effects, central bank purchases).
+- Financing costs (repo rates for the short leg) affect the net P&L of the strategy.