Test Japanese Equity Momentum with a Public Sample

Price momentum turns the direction of price movement over a defined window into a value for comparing securities. A high value is neither an order nor future profit. A backtest also needs rules for turning ranks into holdings and assumptions about when and at which price trades occur.

This page examines the fixed public teaching example. It does not recommend securities, expose a user's strategy or search for the best window. See what a factor is for the distinction between a feature and its ranking.

The public reference uses raw daily bars for 4502, 6758 and 8306 over 1,464 observed sessions from January 4, 2017 to December 30, 2022. The securities and period were fixed beforehand for the lesson. Select, validate and freeze your own local input through shared preparation.

The settings are 20 observed sessions of momentum, monthly rebalancing, the top two holdings, JPY 2 million initial cash and a 0.1% one-way fee. Do not replace observed sessions with calendar days. Record the manifest, settings and software version; keeping conditions unchanged after viewing results is the first assignment.

Create a new research strategy in My Strategies and choose Price momentum (downloaded data). Select the frozen input, enter the stated settings and pass preflight. After completion, check in Backtest Results that the settings and input are associated with this result.

See the existing momentum and indicators tutorial for screen navigation and related indicators. Displaying an indicator separately differs from changing the strategy's order rules. This task follows the calculations and records of one fixed rule.

The lesson forms signals from closes available by the previous observed session and fixes quantities using the prior close. It fills at the actual open of the next observed session, processing same-day sells before buys. Check signal and execution dates, then distinguish the planning price from the actual fill price.

If an opening price change causes insufficient cash, the whole order is skipped rather than conveniently resizing the quantity. Assets are valued at raw closes. The availability and lookahead perspective also requires keeping a subsequently observed opening price out of the previous day's sizing decision.

The public reference has 161 filled orders and 22 skips. Use those counts for reconciliation only when input, recipe and report identities match. If different data produces different counts, do not remove input rows to match the lesson.

The 161 count describes filled orders, not 161 completed trades. Multiple orders can contribute to a holding's opening and closing cycle. Separate prior-day signals, opening fills and skips on the price chart, then inspect reasons and cash changes using reading backtest results.

The lesson excludes dividends, slippage, market impact and board-lot constraints. A 0.1% one-way fee does not represent every real-world trading cost. See transaction costs for the distinctions.

Results are research-only with pit_guarantee=false. Historical availability is uncertified, and this is not a fresh unused evaluation period. Record your ability to explain order timing and skips as the outcome; an equity curve alone cannot establish future effectiveness or readiness for paper observation.

Source: the public GUI research courses. This article teaches research procedures; it does not recommend securities or trades or guarantee future returns. See the disclaimer.

Explore the strategy concepts learning path

• What is momentum? Separate a price trend from a ranking
• What is a moving average? Check its window and availability
• Value investing: separate valuation measures from future returns