Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
16,761 documents
The discussion surveys ways to rank candidate variables for factor models when correlation among inputs is not the main concern. Suggested measures include rank correlation, causal or cointegration tests, information ratios, regression fit, premium…
The document discusses how to handle missing observations in short-term interbank rate data, motivated by a goal of predicting customer behavior. Its suggested first step is to seek a more complete, higher-quality source and use it to repair gaps where…
This discussion asks how practitioners test whether Value at Risk (VaR) violations are independent over time, with particular interest in detecting clustering and deciding when a model needs review. It points to violation-clustering tests in a thesis on VaR…
The document asks how to test whether a strategy’s Sharpe ratio differs significantly from another strategy and how to obtain a p-value. Its response emphasizes that strategy returns are often serially correlated, so ordinary resampling of individual…
The document questions whether linear correlation is sufficient when studying relationships among financial variables and combining volatility forecasts. It describes a proposed comparison of GARCH, support vector regression, and fractionally differenced…
The document clarifies how researchers estimate univariate heterogeneous autoregressive models for realized variance across multiple stocks. The cited study does not pool all stocks into one long dataset to estimate a single common set of coefficients.…
The document distinguishes forecasting a security’s daily high and low from deciding how to trade on those forecasts. The author reports using an ensemble of forecasting methods and evaluating historical errors, including their dispersion and frequency of…
The document answers which sensitivities to use when estimating historical VaR with a delta-gamma approximation. For a portfolio containing an option, the method applies the option's current delta and gamma to a series of historical underlying price changes.…
The document proposes detecting temporal leakage by changing all input rows after a chosen cutoff and rerunning feature construction. If outputs at or before the cutoff change, the feature pipeline has used future data. This empirical check can reveal…
The document shows how to extend an R optimization setup from one parameter to several parameters while continuing to use the optimx package and the L-BFGS-B method. The key change is to represent the starting values as a vector and write the objective…
The document asks how trading firms generate research ideas before examining data, and whether they rely heavily on data mining. The question mentions arbitrage, pairs, mean reversion, tick data, momentum, and factor models, while noting practical data…
The document describes portfolio selection when an investor must choose exactly a limited number of assets from a much larger universe. It frames the objective as maximizing expected return with long or short weights, a gross-exposure constraint, and linear…
The document discusses how to assess whether a sequence of trades in one stock added value through market timing. It notes that fund-level timing models, including Treynor–Mazuy and Merton–Henriksson, may not translate directly to an individual trading…
The document asks how to simulate implied volatility (IV) together with stock prices in a Monte Carlo option analysis. The author generates stock paths with Geometric Brownian Motion using estimated daily return mean and standard deviation, then values…
The document asks why an active strategy’s portfolio value can appear to exceed a passive strategy’s while its calculated returns and cumulative returns look weaker. It provides monthly portfolio value series and shows discrete returns computed as the…
A portfolio sort groups assets by a characteristic, such as expected return, and compares the subsequent performance of the groups. Researchers commonly set periodic breakpoints from the cross-sectional distribution of the chosen variable, then form…
The document explains why a Monte Carlo estimate of a European option’s price will generally differ from its Black–Scholes analytical value, even when the simulation is implemented correctly. Monte Carlo pricing estimates an expected payoff from a finite…
This template trades crossovers between a fast and a slow exponential moving average, with configurable lengths and an optional filter based on the direction of a longer simple moving average. A bullish crossover can open a long when the filter is rising; a…
This XAUUSD-focused strategy records the low during the final hour of the New York session as a reference level, then compares it with lows during the following Asia and London sessions. If Asia trades below that New York low, the system signals a short at…
This strategy identifies a four-swing ABCD harmonic pattern using confirmed pivot highs and lows. It compares the retracement and extension between successive swing prices with Fibonacci-based ratio ranges, then looks for a close between nearby swing levels…
The discussion considers how an opaque trading service might produce buy and sell signals. It offers no verified description of the service’s algorithm: its own marketing claims only that the signals come from proprietary mathematical logic. The answer…
This long-only strategy for ONEUSDT seeks rebounds after a sharp decline. Its entry filters require a specified two-bar price move from five to three bars ago to fall within a decline range, while the latest two-bar move must be flat or negative. The…
This intraday strategy trades opening gaps in SPY, using SPY price data even when applied to another chart. It flags upward or downward gaps when the separation between the current bar and the previous bar’s range exceeds a threshold tied to a configurable…
This strategy defines a rolling range from recent highs and lows, then enters long when a bar closes above the range or short when it closes below. Volume confirmation is enabled by default: the breakout bar’s volume must exceed its moving average, though…