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.
765 documents
The document examines unexpectedly small cointegration-test p-values obtained by applying a test to log returns from two crypto markets. One response highlights a data-quality problem: equal-length candle arrays are not necessarily synchronized. If an…
The document describes frequency-domain methods used to study financial time series, emphasizing wavelets over conventional Fourier analysis. Wavelet methods can examine changing correlations across timescales, estimate time-varying variance as a…
The document examines an apparent contradiction: two identical integrated time series seem to share a stationary relationship, yet a cointegration test may report no cointegration. It explains that the residual from an Engle–Granger regression of one series…
The document examines whether predictability in a stationary process implies mean reversion or momentum, focusing on how different papers use a predictability ratio to select portfolios. It explains a Box–Tiao decomposition that separates total variation…
The document addresses how to interpret coefficients after applying the Johansen cointegration test in a linear time-series setting. The answer advises using the original variables rather than truncating or otherwise altering them for interpretation. It also…
The document presents two Johansen trace test outputs from a pair trading model, using one and two lags, and asks how to interpret them. In both cases, the reported statistics are below their critical values and the p-values exceed the stated significance…
The document explains how to interpret a hedge ratio estimated by regressing one stock’s log price on another’s. The ratio describes the relative percentage sensitivity of the two stocks: for example, a ratio of two means that a one percent logarithmic move…
The document considers how to project future profit and loss and drawdowns for trading strategies and portfolios. It raises several possible approaches: extrapolating a backtest, resampling its empirical returns, forecasting time-varying variance with GARCH…
The document asks how to calculate returns for a spread that can take negative values, since the logarithm of a negative spread level is undefined. It considers adding a constant large enough to make every observation positive, which keeps absolute…
The document considers using a multiple linear regression of a stock’s price on market, commodity, competitor, and currency variables to estimate a model-implied price. The proposed trade goes long when the observed price is below that estimate, with a…
The discussion considers whether reversing one currency quote can make two negatively correlated exchange rates suitable for a cointegration or distance based strategy. It uses EUR/USD and USD/JPY as examples, noting that quote direction changes the way…
In rates trading, an outright position is a trade in an instrument taken on its own, rather than as one leg of a broader paired or hedging structure. When a trader requests a price for a swap, a dealer may ask whether the trade is outright to understand…
The document asks how to translate a pairs-trading spread signal into positions in two assets. It defines a spread as one price minus beta times the other and considers entry signals when the spread moves above or below its mean by a standard deviation. The…
Pairs trading is presented as a relative-value position: take a long position in an asset expected to outperform and a short position in a related asset. A stock example pairs a long position in General Motors with a short position in Ford. The intended gain…
The document explains how to read output from Johansen and Engle–Granger cointegration tests for a pair of time series. In the example, the test indicators fail to reject their null hypotheses, and the reported p-values are high. The answer interprets this…
The document asks how to express forecasts of relative stock performance in a mean-variance portfolio optimizer for a medium-frequency strategy with holding periods of about a day to a week. It frames statistical arbitrage as trading relationships among…
The document explains why a statistically cointegrated pair can still perform poorly in a trading backtest. It identifies transaction costs as a major obstacle, especially when the spread between related securities is small, and warns that statistical mean…
The document asks how to extend a Kalman filter regression commonly used to estimate a rolling spread between two asset price series. The proposed relationship models one asset’s price as a linear combination of two other asset prices, with an optional…
The document discusses how to extend pairs trading to baskets of more than two securities. Its central idea is to seek portfolio weights that produce a stationary portfolio series, then trade deviations as they revert. It points to general portfolio methods…
The document addresses how to represent a persistent, nonzero equilibrium spread between prices that are cointegrated, such as the same instrument traded in different markets. In a bivariate vector error-correction model, the proposed approach includes a…
The document asks how to chart a long-short position when its two stocks have different capital weights or market sensitivities. One response says to combine each stock’s price or profit-and-loss contribution using the same weights as the position, so the…
The document asks how to turn a factor that tracks price into buy and sell decisions, using the difference between German and US ten-year yields alongside EUR/USD as an example. One response describes a pairs trading approach: first test whether the series…
The document examines a common pairs-trading heuristic: regress each asset on the other, construct a spread from each regression, and choose the hedge ratio associated with the stronger Augmented Dickey-Fuller result. It argues that this choice lacks a sound…
The document surveys ways to compare stock time series, with the practical suggestion to work with returns or differences rather than trending price levels. It reviews Pearson correlation for linear association and rank-based Spearman and Kendall measures…