Data to 20 November 2026
About

Model library · Data to 20 Nov 2026

The rules, their holdings and their records

See what the rules hold, why they differ and what their records show. Start with 17 representative models; explore all 61 rules and variants.

Historical research, not investment advice. Everything before a model entered the library is a backtest.

The latest allocation targets

Representative rules, chosen by construction. Targets use the decision dates below and trade at the next close. A target dated today may still await execution; a partial week can change before the weekly close. Provisional monthly signals are shown separately.

Simple allocations

60/20/20

Decision 31 Dec 2025 · India only

60% Nifty 500, 20% G-secs, 20% gold

Equity 60%Debt 20%Gold 20%

Simple allocations

60/40

Decision 31 Dec 2025 · India only

60% Nifty 500, 40% G-secs

Equity 60%Debt 40%

Simple allocations

Equal thirds

Decision 31 Dec 2025 · India only

33% gold, 33% Nifty 500, 33% G-secs

Equity 33%Debt 33%Gold 33%

Simple allocations

Permanent portfolio

Decision 31 Dec 2025 · India only

25% Nifty 500, 25% long G-secs, 25% gold, 25% cash

Equity 25%Debt 25%Gold 25%Cash 25%

Simple allocations

1/N, India

Decision 30 Oct 2026 · India only

20% Nifty 100, 20% Midcap 150, 20% Smallcap 250, 20% gold, 20% G-secs

Equity 60%Debt 20%Gold 20%

Trend

Nifty 500, 10-month rule

Decision 30 Oct 2026 · India only

Cash

Cash 100%

If the month ended on 20 Nov 2026

Nifty 500

The current target remains in force until the next completed decision is executed.

Trend

Trend, India only

Decision 20 Nov 2026 · India only

16% Nifty 100, 14% Smallcap 250, 13% Midcap 150, 6% G-secs, 52% cash

Equity 43%Debt 6%Cash 52%

Trend

Trend, global equity, ACWI (ETF version)

Decision 20 Nov 2026 · Includes foreign assets

13% Nifty 100, 12% Smallcap 250, 10% Midcap 150, 5% G-secs, 3% Global equity (ACWI), 57% cash

Equity 38%Debt 5%Cash 57%

Momentum and rotation

Dual momentum, India only

Decision 30 Oct 2026 · India only

Gold

Gold 100%

If the month ended on 20 Nov 2026

G-secs

The current target remains in force until the next completed decision is executed.

Momentum and rotation

Size rotation

Decision 30 Oct 2026 · India only

G-secs

Debt 100%

Momentum and rotation

Sector rotation

Decision 30 Oct 2026 · India only

33% Metal, 33% Healthcare, 33% cash

Equity 67%Cash 33%

If the month ended on 20 Nov 2026

33% Metal, 33% Media, 33% Healthcare

The current target remains in force until the next completed decision is executed.

Momentum and rotation

Factor rotation

Decision 30 Oct 2026 · India only

Cash

Cash 100%

Valuation and market size

Valuation glide (CAPE)

Decision 30 Oct 2026 · India only

64% G-secs, 36% Nifty 500

Equity 36%Debt 64%

Valuation and market size

Countercyclical indexing

Decision 30 Oct 2026 · India only

62% G-secs, 38% Nifty 500

Equity 38%Debt 62%

Risk allocation

Inverse volatility

Decision 30 Oct 2026 · India only

78% G-secs, 14% Nifty 500, 7% gold

Equity 14%Debt 78%Gold 7%

Other tactical rules

GTAA, India only

Decision 30 Oct 2026 · India only

33% G-secs, 67% cash

Debt 33%Cash 67%

If the month ended on 20 Nov 2026

33% Nifty 500, 33% G-secs, 33% cash

The current target remains in force until the next completed decision is executed.

Monthly rules decide at month-end and trade at the next close. Weekly and daily rules follow their stated schedule. Find every version or view the published index series.

Different rules, different jobs

A variation on the same signal is useful research, but it may add little diversification. Models are grouped by what changes their allocation. Each starting point is followed by its variants and related rules.

61 models and variants

Simple allocations 12Own a fixed mix and rebalance it.

1/N, IndiaStarting point

Equal fifths in the Nifty 100, Nifty Midcap 150, Nifty Smallcap 250, gold and the 5-year G-sec, put back to equal weights at every month-end.

Backtest May 2005 to Nov 2026. Entered 14 Oct 2026.

60/20/20Starting point

60% Nifty 500 TRI, 20% 5-year G-sec, 20% gold, rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

60/40Starting point

60% Nifty 500 TRI, 40% 5-year G-sec, rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Permanent portfolioStarting point

25% each in Nifty 500 TRI, long G-secs (synthetic 10-year), gold and cash (91-day T-bill), rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Equal thirdsStarting point

One third each in Nifty 500 TRI, 5-year G-sec and gold, rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

1/N, with global

Equal sixths in the Nifty 100, Nifty Midcap 150, Nifty Smallcap 250, gold, the 5-year G-sec and the S&P 500 in rupees (through LRS), put back to equal weights at every month-end.

Backtest May 2005 to Nov 2026. Entered 14 Oct 2026.

Golden butterfly

20% each in Nifty 500 TRI, Nifty Smallcap 250 TRI, long G-secs (synthetic 10-year), cash and gold, rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Endowment-style (Indian approximation)

30% Nifty 500, 10% Nifty Midcap 150, 10% Nifty Smallcap 250, 20% gold, 15% long G-secs (synthetic 10-year), 15% 5-year G-sec, rebalanced each January.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Equity and debt, 65/35

65% Nifty 500 and 35% 5-year G-sec, rebalanced to target at the end of March and September, when NSE's broad indices are reviewed.

Backtest Apr 2005 to Nov 2026. Entered 15 Oct 2026.

Equity and debt, 50/50

50% Nifty 500 and 50% 5-year G-sec, rebalanced to target at the end of March and September, when NSE's broad indices are reviewed.

Backtest Apr 2005 to Nov 2026. Entered 15 Oct 2026.

Equity and debt, 25/75

25% Nifty 500 and 75% 5-year G-sec, rebalanced to target at the end of March and September, when NSE's broad indices are reviewed.

Backtest Apr 2005 to Nov 2026. Entered 15 Oct 2026.

Trend 14Reduce exposure when a market weakens.

Trend, global equity, ACWI (ETF version)Starting point

Six equal sleeves: Nifty 100, Midcap 150, Smallcap 250, gold, 5-year G-secs and developed and emerging global equity through iShares ACWI. Exposure in each sleeve equals its 24-vote trend score; the rest is cash. Weekly decisions with a 5-percentage-point trade band.

Backtest Apr 2009 to Nov 2026. Entered 17 Oct 2026.

Trend, India onlyStarting point

Five equal sleeves: Nifty 100, Nifty Midcap 150, Nifty Smallcap 250 (together, the Nifty 500 with no overlap), gold and the 5-year G-sec. Each sleeve is invested in proportion to its trend score (24 votes, 0 to 100) and holds cash for the rest. Decided each week; a sleeve trades only when it is more than 5 points off its target.

Backtest Apr 2006 to Nov 2026. Entered 8 Oct 2026.

Nifty 500, 10-month ruleStarting point

Hold the Nifty 500 when its month-end level is above the average of the last 10 month-ends; otherwise cash.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

Trend, India only (first version, with the Nifty 50)

Five equal sleeves: Nifty 50, Nifty Midcap 150, Nifty Smallcap 250, gold and the 5-year G-sec. Each sleeve is invested in proportion to its trend score (24 votes, 0 to 100) and holds cash for the rest. Decided each week; a sleeve trades only when it is more than 5 points off its target.

Backtest Apr 2006 to Nov 2026. Entered 8 Oct 2026.

Nifty 50, graded by trend score

Nifty 50 exposure equals its trend score divided by 100, the rest in cash. Decided weekly; trades only when more than 5 points off target.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

Nifty 50, 10-month rule

Hold the Nifty 50 when its month-end level is above the average of the last 10 month-ends; otherwise cash.

Backtest Apr 2005 to Nov 2026. Entered 1 Oct 2026.

Nifty 50, 200-day rule

Hold the Nifty 50 when it closes above its 200-session average; otherwise cash. Checked every session.

Backtest Apr 2005 to Nov 2026. Entered 1 Oct 2026.

Trend, US large caps, IVV (ETF version)

Six equal sleeves: Nifty 100, Midcap 150, Smallcap 250, gold, 5-year G-secs and US large caps through iShares IVV. Exposure in each sleeve equals its 24-vote trend score; the rest is cash. Weekly decisions with a 5-percentage-point trade band.

Backtest Apr 2006 to Nov 2026. Entered 17 Oct 2026.

Trend, with global

As Trend, India only, with a sixth sleeve for the S&P 500 in rupees held through LRS: six equal sleeves.

Backtest Apr 2006 to Nov 2026. Entered 8 Oct 2026.

Robust trend

Five equal sleeves (Nifty 100, Midcap 150, Smallcap 250, gold, 5-year G-sec). Each month-end a sleeve is half invested if its level is above its 12-month average and half if its 12-month return beats cash; the rest of the sleeve is in cash.

Backtest May 2006 to Nov 2026. Entered 14 Oct 2026.

Momentum and rotation 15Choose markets with stronger past returns.

Dual momentum, India onlyStarting point

Hold whichever of Nifty 500 or gold had the higher 12-month return, if it beat cash; otherwise the 5-year G-sec.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Dual momentum, global equity, ACWI (ETF version)Starting point

If the Nifty 500 beats cash over 12 months, hold whichever of Nifty 500 and developed and emerging global equity through iShares ACWI has the stronger 12-month return; otherwise hold 5-year G-secs.

Backtest Apr 2009 to Nov 2026. Entered 17 Oct 2026.

Sector rotationStarting point

As the top-3 rotation, but a slot goes to cash when that sector's return over months t-12 to t-1 is below cash's.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

Size rotationStarting point

If the Nifty 500's 12-month return beats cash, hold whichever of the Nifty 100, Midcap 150 and Smallcap 250 (together, the Nifty 500) did best over 12 months; otherwise hold the 5-year G-sec.

Backtest May 2006 to Nov 2026. Entered 8 Oct 2026.

Factor rotationStarting point

Each month, hold equally the two of four factor indices (momentum, quality, value, low volatility) with the best return over months t-12 to t-1; a slot goes to cash when its factor's return is below cash's.

Backtest May 2006 to Nov 2026. Entered 8 Oct 2026.

Dual momentum, with the S&P 500

If the Nifty 500's 12-month return beats cash, hold whichever of the Nifty 500 and the S&P 500 in rupees (through LRS) did better over 12 months; otherwise hold the 5-year G-sec.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

Sector rotation (sectors counted from launch)

Each month, hold equally the three of twelve sectors with the best return over months t-12 to t-1; a slot goes to cash when its sector's return is below cash's. Each sector counts only from 13 months after NSE launched it.

Backtest Aug 2012 to Nov 2026. Entered 8 Oct 2026.

Factor rotation (factors counted from launch)

Each month, hold equally the two of four factor indices (Nifty200 Momentum 30, Nifty200 Quality 30, Nifty500 Value 50, Nifty100 Low Volatility 30) with the best return over months t-12 to t-1; a slot goes to cash when its factor's return is below cash's. Each factor counts only from 13 months after NSE launched it; the rule runs from the first month with three.

Backtest Nov 2019 to Nov 2026. Entered 8 Oct 2026.

Robust equity momentum

Each week, eight rules vote: for lookbacks of 3, 6, 9 and 12 months (63 to 252 sessions), the best of Nifty 100, Midcap 150, Smallcap 250 and the S&P 500 in rupees gets the vote if it beat cash (four rules) or closed above its average (four rules) over that lookback; otherwise the vote goes to the 5-year G-sec. Weights are the vote shares; trades only when a weight is more than 5 points off target.

Backtest Apr 2006 to Nov 2026. Entered 14 Oct 2026.

Adaptive asset allocation

Each month-end, the three of seven assets (Nifty 100, Midcap 150, Smallcap 250, gold, 5-year G-sec, 10-year G-sec, S&P 500 in rupees) with the best six-month return, weighted for minimum variance on the last 60 sessions' covariance (Ledoit-Wolf shrinkage).

Backtest Oct 2005 to Nov 2026. Entered 14 Oct 2026.

Valuation and market size 4Change equity exposure with prices or market size.

Countercyclical indexingStarting point

Each month-end, equity's share of equity plus government debt is the market value of listed companies over the sum of that and government debt (both as a % of GDP; debt as BIS published it). Hold the inverse: that share in the 5-year G-sec and the rest in the Nifty 500. Trades only when a weight is more than 5 points off target.

Backtest Apr 2005 to Nov 2026. Entered 14 Oct 2026.

Valuation glide (CAPE)Starting point

Equity share moves from 80% when the Sensex CAPE10 (IIMA) is at or below its expanding 20th percentile to 20% at or above the 80th; the rest in the 5-year G-sec; one-month lag.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Dynamic allocation

Each month, Nifty 500 equity at the average of a valuation half (80% when its P/E is in its cheapest fifth since 1999, falling to 20% in its dearest fifth) and a trend half (20% at a trend score of 0, rising to 80% at 100); the rest in the 5-year G-sec. Trades only when a weight is more than 5 points off target.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

Valuation glide (yield gap)

Equity share moves from 80% when the Nifty 500 earnings yield minus the 10-year G-sec yield is at or above its expanding 80th percentile (cheap) to 20% at or below the 20th; the rest in the 5-year G-sec; one-month lag.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Risk allocation 5Weight assets by their measured risk.

Inverse volatilityStarting point

Weights proportional to 1/volatility over the last 252 sessions; Nifty 500, 5-year G-sec, gold; monthly.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Minimum variance

Long-only minimum variance on the Ledoit-Wolf shrunk covariance of the last 252 sessions; monthly.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Hierarchical risk parity

Single-linkage clustering on correlation distance, then recursive bisection with inverse-variance weights; last 252 sessions; monthly.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Maximum diversification

Long-only weights that maximise the diversification ratio over the last 252 sessions; monthly.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Other tactical rules 11Combine allocation rules or defensive signals.

GTAA, India onlyStarting point

One third each to Nifty 500, 5-year G-sec and gold; a sleeve whose month-end level is below its 10-month average sits in cash.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

GTAA, with global

Four equal sleeves: Nifty 500, S&P 500 in rupees (through LRS), the 5-year G-sec and gold. Each is held when its month-end level is above its 10-month average, otherwise that sleeve is in cash.

Backtest Apr 2005 to Nov 2026. Entered 8 Oct 2026.

GTAA, US large caps, IVV (ETF version)

Four equal sleeves: Nifty 500, US large caps through iShares IVV, 5-year G-secs and gold. Hold a sleeve above its 10-month average at month-end; otherwise hold cash.

Backtest Apr 2005 to Nov 2026. Entered 17 Oct 2026.

GTAA, global equity, ACWI (ETF version)

Four equal sleeves: Nifty 500, developed and emerging global equity through iShares ACWI, 5-year G-secs and gold. Hold a sleeve above its 10-month average at month-end; otherwise hold cash.

Backtest Jan 2009 to Nov 2026. Entered 17 Oct 2026.

Protective asset allocation (PAA2)

Momentum = price / 12-month average - 1 on Nifty 50, Next 50, Midcap 100, Smallcap 100 and gold. The bond share is (N - n)/(N - N/2) where n assets have positive momentum; the rest goes equally to the top 3 with positive momentum; bonds are the 5-year G-sec.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Defensive asset allocation (DAA)

Canaries Smallcap 100 and the 5-year G-sec: each with negative 13612W momentum moves half the portfolio to the better of G-sec and cash. The rest goes equally to the top 2 of Nifty 50, Next 50, Midcap 100 and gold by 13612W.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Vigilant asset allocation (VAA-G4)

13612W momentum. If Nifty 50, Midcap 100, Smallcap 100 and gold are all positive, hold the best of them; otherwise the better of the 5-year G-sec and cash.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Bold asset allocation (BAA-G4)

If any canary (Nifty 50, Smallcap 100, gold, 5-year G-sec) has negative 13612W momentum, hold the best of 5-year G-sec, 10-year G-sec and cash by price over its 12-month average (cash if it is better); otherwise the top 2 of Nifty 50, Next 50, Midcap 100, Smallcap 100 and gold by the same measure.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

Volatility-managed equity

Equity share = (median past monthly variance) / (last month's realised variance of Nifty 500), capped at 100%; the rest in cash; monthly.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

60/40 with trend

60/40, with the equity sleeve in cash when Nifty 500 is below its 10-month average.

Backtest Apr 2005 to Nov 2026. Entered 2 Oct 2026.

The India and global versions can change both the assets and the decision rule. Their difference does not isolate the benefit of adding a foreign investment.

Give every model the same test period

Longer histories contain different market conditions. Compare the same dates before interpreting a difference in return or risk.

Choose up to four models. Every line starts on the same date, with one rupee. Compare return with the depth and duration of losses.

Common observations, 2006-04-13 to 2026-11-20. After costs, before tax.

Growth over the same dates

How to read it. Equal slopes on this log scale mean equal rates of growth. All paths start at one rupee.

Trend, India only60/20/20

Growth over the same dates. Line chart, 13 Apr 2006 to 20 Nov 2026. Trend, India only: latest 9.78×, range 0.93× to 10.04×. 60/20/20: latest 11.20×, range 0.89× to 11.55×.

Source: Tipsheet model curves; weekly observations
Method

Falls from a peak over the same dates

How to read it. Zero is a new high. A deeper fall is a larger loss; a longer stretch below zero is a longer wait to recover.

Trend, India only60/20/20

Falls from a peak over the same dates. Line chart, 13 Apr 2006 to 20 Nov 2026. Trend, India only: latest −2.5%, range −16.4% to 0.0%. 60/20/20: latest −3.1%, range −36.1% to 0.0%.

Source: Tipsheet model curves; weekly observations
Method
Common-window results, after costs and before tax. Falls measured on weekly observations.
ModelReturn a yearWorst weekly-observed fall
Trend, India only11.7%−16.4%
60/20/2012.4%−36.1%

These are historical backtests, including periods before the rules entered this library. Returns are measured between shared observations, with no missing values filled. Weekly observations can miss a deeper daily loss. Tax needs a fresh simulation for each start date, so these comparisons are before tax. Each model page states its full-record tax basis; ETF versions do not yet model investor tax. A registered benchmark can have different risk or different assets; read its definition.

What the evidence supports

6 of 44 tax-modelled rules finished ahead of their registered benchmark after tax. None has a before-tax annual return gap whose 90% uncertainty interval lies wholly above zero. These are historical comparisons; a defensive rule can sacrifice return to reduce losses.

Read the right comparison

A same-assets benchmark tests allocation decisions against a simpler mix. A broad-market benchmark also changes the asset mix. Each model page states the actual benchmark.

Return, loss and implementation are separate questions. An uncertain return gap does not establish that two strategies are equivalent.

What the figures include

Full-record returns below include model costs. Each row states its tax basis: ETF versions are before investor tax and dividend withholding; older versions retain their stated Indian-tax assumptions. Worst falls and uncertainty intervals are before tax. The Compare view uses weekly observations, after costs and before tax.

The ETF versions use IVV for US large caps or ACWI for developed and emerging global equity, including distributions. The older S&P 500 and Nasdaq price-index versions remain available with their dividends excluded. ETF histories are not extended before inception. NAV returns already include fund expenses; assumed execution costs are additional.

A backtest and a recorded history are different

Entry into the library fixes a rule. It does not turn the earlier history into a live record. A session filled after its date is reconstructed; returns recorded on the date form the contemporaneous paper record. Neither is an executed portfolio.

60/20/2033 logged sessions; 33 reconstructed; log through 19 Nov 2026.

60/4033 logged sessions; 27 reconstructed; log through 19 Nov 2026.

Equal thirds33 logged sessions; 33 reconstructed; log through 19 Nov 2026.

Permanent portfolio33 logged sessions; 33 reconstructed; log through 19 Nov 2026.

1/N, India26 logged sessions; 26 reconstructed; log through 19 Nov 2026.

Nifty 500, 10-month rule31 logged sessions; 15 reconstructed; log through 20 Nov 2026.

Trend, India only30 logged sessions; 30 reconstructed; log through 19 Nov 2026.

Trend, global equity, ACWI (ETF version)23 logged sessions; 23 reconstructed; log through 19 Nov 2026.

Dual momentum, India only33 logged sessions; 33 reconstructed; log through 19 Nov 2026.

Dual momentum, global equity, ACWI (ETF version)23 logged sessions; 19 reconstructed; log through 19 Nov 2026.

Size rotation30 logged sessions; 24 reconstructed; log through 19 Nov 2026.

Sector rotation31 logged sessions; 15 reconstructed; log through 20 Nov 2026.

Factor rotation31 logged sessions; 15 reconstructed; log through 20 Nov 2026.

Valuation glide (CAPE)33 logged sessions; 27 reconstructed; log through 19 Nov 2026.

Countercyclical indexing26 logged sessions; 20 reconstructed; log through 19 Nov 2026.

Inverse volatility33 logged sessions; 33 reconstructed; log through 19 Nov 2026.

GTAA, India only33 logged sessions; 33 reconstructed; log through 19 Nov 2026.

Full records, with their own dates

This is a research inventory. Use Compare for a common-window comparison. Annual returns and benchmark gaps use the tax basis stated in each row; worst fall and the 90% interval are before tax. Benchmark gaps are percentage points a year.

Each model over its own dates. Multiple testing counts 203 trials. A range spanning zero leaves the return gap uncertain.

Each model over its own dates. Multiple testing counts 203 trials. A range spanning zero leaves the return gap uncertain.
ModelFromToTax basisReturn a yearWorst daily fall, before taxBenchmark gap90% interval, before taxMore
Trend, India only (first version, with the Nifty 50)Apr 2006Nov 2026After tax9.4%−17.6%−3.6−4.6 to 0.4
Trend, with global (first version, with the Nifty 50)Apr 2006Nov 2026After tax8.5%−16.1%−4.5−5.3 to −0.5
Nifty 50, graded by trend scoreApr 2005Nov 2026After tax9.3%−25.0%−3.4−6.1 to 1.9
Nifty 50, 10-month ruleApr 2005Nov 2026After tax12.3%−29.8%−0.4−4.4 to 4.5
Nifty 50, 200-day ruleApr 2005Nov 2026After tax10.8%−29.9%−2.0−5.8 to 2.8
Nifty 50, 1/3/6/12-month blendApr 2005Nov 2026After tax9.3%−26.9%−3.4−6.5 to 1.5
Midcap 150, 200-day ruleJan 2006Nov 2026After tax14.8%−37.7%0.7−4.9 to 7.9
Smallcap 250, 200-day ruleJan 2006Nov 2026After tax14.0%−36.9%1.7−5.8 to 9.2
GTAA, India onlyApr 2005Nov 2026After tax9.8%−18.0%−2.4−3.2 to −0.8
GTAA, with globalApr 2005Nov 2026After tax9.4%−14.8%−2.8−3.3 to −0.7
Dual momentum, India onlyApr 2005Nov 2026After tax11.3%−45.3%−0.9−5.0 to 4.7
Dual momentum, with the S&P 500Apr 2005Nov 2026After tax9.9%−45.1%−3.4−6.5 to 2.8
Dual momentum, with the Nasdaq-100Apr 2005Nov 2026After tax8.5%−45.1%−7.3−11.2 to −0.2
Trend, US large caps, IVV (ETF version)Apr 2006Nov 2026Before investor tax11.0%−16.1%−3.1−5.3 to −0.5
GTAA, US large caps, IVV (ETF version)Apr 2005Nov 2026Before investor tax11.2%−14.7%−2.3−3.4 to −1.0
Dual momentum, US large caps, IVV (ETF version)Apr 2005Nov 2026Before investor tax12.2%−45.1%−2.8−7.5 to 2.2
Trend, global equity, ACWI (ETF version)Apr 2009Nov 2026Before investor tax11.1%−9.2%−4.0−5.8 to −2.1
GTAA, global equity, ACWI (ETF version)Jan 2009Nov 2026Before investor tax10.2%−10.7%−2.7−3.9 to −1.6
Dual momentum, global equity, ACWI (ETF version)Apr 2009Nov 2026Before investor tax10.1%−19.0%−5.7−9.7 to −1.9
Size rotation (first version, with the Nifty 50)May 2006Nov 2026After tax8.6%−49.1%−3.3−8.6 to 3.3
Sector rotation, top 3 (NSE history)Apr 2005Nov 2026After tax11.3%−66.0%−2.4−5.9 to 2.5
Sector rotationApr 2005Nov 2026After tax10.3%−54.4%−3.3−8.1 to 2.4
Sector rotation, top 3 (sectors counted from launch)Aug 2012Nov 2026After tax12.4%−38.6%0.2−2.5 to 4.8
Sector rotation (sectors counted from launch)Aug 2012Nov 2026After tax9.8%−37.4%−2.4−6.9 to 2.9
Trend, India onlyApr 2006Nov 2026After tax9.5%−17.6%−3.5−4.5 to 0.5
Trend, with globalApr 2006Nov 2026After tax8.5%−16.1%−4.6−5.3 to −0.5
Nifty 500, 10-month ruleApr 2005Nov 2026After tax11.2%−33.9%−1.8−6.3 to 3.9
Size rotationMay 2006Nov 2026After tax8.6%−49.1%−3.4−8.9 to 3.2
Factor rotationMay 2006Nov 2026After tax8.9%−51.5%−4.8−9.7 to 0.0
Factor rotation (factors counted from launch)Nov 2019Nov 2026After tax7.2%−29.1%−6.8−15.1 to 0.2
Dynamic allocationApr 2005Nov 2026After tax10.2%−32.4%−0.8−1.0 to 0.4
1/N, IndiaMay 2005Nov 2026After tax13.8%−45.5%0.4−2.6 to 3.7
1/N, with globalMay 2005Nov 2026After tax13.5%−43.0%0.2−3.3 to 3.9
Robust trendMay 2006Nov 2026After tax9.2%−25.8%−3.6−6.5 to −0.9
Robust equity momentumApr 2006Nov 2026After tax7.5%−42.3%−5.5−8.7 to 1.5
Adaptive asset allocationOct 2005Nov 2026After tax8.5%−32.8%−4.2−6.8 to 0.8
Countercyclical indexingApr 2005Nov 2026After tax11.2%−31.6%0.2−0.7 to 0.9
Risk parity (equal risk contribution)Apr 2005Nov 2026After tax8.8%−8.0%−3.4−5.5 to −1.5
Minimum varianceApr 2005Nov 2026After tax7.6%−5.9%−4.6−7.4 to −2.0
Inverse volatilityApr 2005Nov 2026After tax8.8%−7.2%−3.3−5.5 to −1.4
Hierarchical risk parityApr 2005Nov 2026After tax7.2%−6.3%−5.0−8.2 to −2.4
Maximum diversificationApr 2005Nov 2026After tax8.8%−8.9%−3.4−5.4 to −1.5
All Weather (Indian approximation)Apr 2005Nov 2026After tax10.1%−16.2%––
Golden butterflyApr 2005Nov 2026After tax12.3%−23.7%––
Endowment-style (Indian approximation)Apr 2005Nov 2026After tax13.2%−30.7%––
Valuation glide (CAPE)Apr 2005Nov 2026After tax10.2%−15.7%−0.8−2.4 to 1.7
Valuation glide (yield gap)Apr 2005Nov 2026After tax10.4%−29.1%−0.6−1.6 to 1.7
Protective asset allocation (PAA2)Apr 2005Nov 2026After tax12.4%−36.6%−1.9−4.7 to 2.1
Defensive asset allocation (DAA)Apr 2005Nov 2026After tax12.7%−45.1%−1.6−3.7 to 2.9
Vigilant asset allocation (VAA-G4)Apr 2005Nov 2026After tax10.0%−31.2%−4.0−6.5 to 1.6
Bold asset allocation (BAA-G4)Apr 2005Nov 2026After tax8.9%−35.2%−4.5−7.5 to 0.8
Volatility-managed equityApr 2005Nov 2026After tax10.4%−46.4%−2.6−5.0 to 0.1
60/40 with trendApr 2005Nov 2026After tax9.7%−21.0%−1.3−3.0 to 0.7
Ensemble: equal weight of modelsApr 2005Nov 2026After tax10.1%−25.1%––
60/20/20Apr 2005Nov 2026After tax13.1%−36.1%––
60/40Apr 2005Nov 2026After tax11.6%−36.5%––
Equity and debt, 65/35Apr 2005Nov 2026After tax11.4%−46.0%––
Equity and debt, 50/50Apr 2005Nov 2026After tax10.5%−36.4%––
Equity and debt, 25/75Apr 2005Nov 2026After tax8.7%−17.6%––
Permanent portfolioApr 2005Nov 2026After tax10.4%−14.7%––
Equal thirdsApr 2005Nov 2026After tax12.2%−18.9%––

Retired experiments stay on the record

  • Sector momentum v1, 12-1 months. Trailed equal weight across the same sectors: 14.5% against 15.5% a year (2005-02 to 2026-09), deflated Sharpe 0.33. NSE's sector list held near-duplicates; v2 fixes the list in advance.
  • Sector momentum v1, 6-1 months. Led equal weight (16.9% against 15.5%) but not significantly: deflated Sharpe 0.66 with 2 trials.

Read the full method, registered rules and results log, trend evidence and portfolio lab. Model detail pages retain their signals, holdings histories and research sources.