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 document describes an attempt to forecast a differenced FTSE 100 series with an ARMA mean model and GARCH disturbances using the rugarch package. The author encounters missing volatility forecasts when requesting rolling forecasts and a plotting error…
The document considers how to assemble historical intraday data for penny stocks that gap up by at least 10%, using a definition based on low share price and market capitalization. Its suggested workflow is to identify qualifying gap-up events with a stock…
The document asks how to calibrate the market price of risk in the Black–Karasinski short-rate model when zero-coupon bond prices lack a closed-form expression. The proposed Monte Carlo approach simulates short-rate paths, discounts cash flows along them,…
The discussion surveys resources and cautions for applying statistical tests to trading ideas. It recommends evidence-based technical analysis as an introduction to objective indicator testing, including statistical inference, Monte Carlo permutation…
The document asks for a stock market anomaly that is straightforward to reproduce in an event study. The motivating constraint is that some candidate anomalies, such as post-earnings-announcement drift, are harder to demonstrate because company events occur…
The document asks how to simulate a multifactor Ornstein–Uhlenbeck process with matrix-valued mean reversion and diffusion. It contrasts the scalar exact time-step formula with the multivariate solution, whose random increment has correlated components.…
The document collects approaches to generating synthetic price series that remain within a chosen range while exhibiting a specified, but imperfect, relationship to historical prices. One answer proposes modeling returns with an Ornstein–Uhlenbeck process…
The document explores whether predictive models should use market closing prices or adjusted prices, which account for distributions such as dividends. The author considers a model forecasting Dow Jones returns from other indices, commodities, and…
The document describes a Monte Carlo exercise procedure sometimes called hindsight overhedge for path-dependent options. At an exercise date, it examines the simulated path through a later maturity and uses that future information to decide retrospectively…
The document discusses the difficulty of using wavelet transforms causally on finite financial time series. Boundary handling methods such as padding, symmetry, polynomial extrapolation, or forecasting an extension can influence coefficients near the…
The response explains how Bayesian modeling could relate strategy performance to changing factors, while correcting the idea that historical observations themselves constitute a prior. It distinguishes observed data from prior information about unknown…
The document raises a practical concern about estimating Value-at-Risk with a circular block bootstrap. The author asks how to wrap sampled return blocks from the end of a series back to its beginning, particularly around the observations labeled P10 and P1,…
The discussion examines an apparent case in which a numerically calculated American call is slightly cheaper than an analytically priced European call under matching inputs. The accepted response attributes the discrepancy to comparing different numerical…
The discussion explains why a high-frequency backtest’s latency assumption should reflect both message timing and the strategy’s sensitivity to event ordering. For spread-crossing or arbitrage strategies, it suggests modeling latency as variable, potentially…
The document poses a research question about whether a stock’s price relative to its 52-week high can help explain subsequent returns. The proposed setup uses the price-to-high ratio as a predictor and a forward return, such as the following week’s return,…
The discussion addresses how to compare a resampled efficient frontier (REF) with standard mean–variance optimization (MVO) when expected returns and a covariance matrix are available but future outcomes are not. It presents two sharply conflicting views of…
The document explains a simple way to generate two non-stationary time series that are cointegrated. Start with a series containing a unit root, such as a random walk, then form a second series by adding a stationary process. The shared stochastic trend…
The document examines an R example that assigns long and short positions from buy and sell signals, carries the latest signal forward, multiplies it by discrete close-to-close returns, and exponentiates the cumulative sum. The replies explain why the…
The document considers whether individual quarterly returns can be recovered from a sequence of rolling one-year returns. Each rolling figure combines several quarterly returns, so the unknown quarterly values can be represented as a system of equations. The…
This discussion collects risk management topics relevant to automated trading, with particular attention to high-frequency systems. It names drawdowns, runs of losses, empirical loss distributions, confidence intervals, value at risk, slippage, liquidity,…
The document asks how a martingale should appear in a simulation of a Cox–Ross–Rubinstein stock-price model. It proposes checking whether the expected ratio of successive prices is one, then wonders whether plotting prices through time would reveal the…
The note addresses whether days with zero observed stock returns should be removed before calculating excess returns in a CAPM analysis. Its main guidance is to first determine whether the zeros represent genuine market observations or data errors. Erroneous…
The document presents a regression question about two predictors whose individual coefficient tests produce relatively large p-values, while a joint test of both coefficients being zero produces a smaller p-value. It asks why this can happen and seeks…
The document asks how to find past expiration dates for futures, especially stock index contracts, and contrasts calculating dates with retrieving recorded contract data. One answer describes querying historical instrument definitions from a market data…