About this module
Financial data has an information clock.
We will distinguish levels from changes, lags from leads, rolling memory from expanding memory, and an event date from its release date.
Every forecast and backtest inherits these timing choices, even when the final model looks mathematically sophisticated.
Lessons
- Lesson 91
Time Order, Frequency, Regularity, and Financial Calendars
A trading calendar is not an evenly spaced stopwatch
2:33 - Lesson 92
Levels, Changes, Differences, and Returns
Separating account levels, cash changes, and relative returns
2:35 - Lesson 93
Lags, Leads, and Temporal Dependence
Keeping past features separate from future labels
2:22 - Lesson 94
Rolling and Expanding Windows
Choosing how much history a monitoring statistic remembers
2:26 - Lesson 95
Resampling, Aggregation, and Time Alignment
Aggregating event payments without double-counting a boundary
2:33 - Lesson 96
Trend, Seasonality, Cycles, and Remainder
Separating a growing business from a repeating weekly pattern
2:38 - Lesson 97
Autocovariance and Autocorrelation
Measuring lagged similarity on a declared finite-sample convention
2:29 - Lesson 98
Stationarity and Differencing Intuition
Removing a changing level without declaring stationarity solved
2:39 - Lesson 99
Smoothing, Baselines, and Naive Forecasts
Giving an advanced forecast a simple baseline to beat
2:31 - Lesson 100
Look-Ahead Leakage and Time-Aware Data Splits
Blocking late information even when its event date is old
2:30