Family 4 of 9 · 10 formulas · each in 3D
Statistics
Learning from data: averages, standard errors, confidence intervals, tests, p-values and likelihood. A backtest is a sample. Statistics tells you how much of its result is skill and how much is luck.
Showing all 10
- 01beginner
Sample mean
Add the observations and divide by how many there are. It is the best simple guess at the true mean, but it moves from sample to sample.
Used by quant researcher: Estimates a strategy's average daily return from its backtest.
Full lesson - 02intermediate
Sample variance (Bessel's correction)
The average squared distance from the sample mean, but divided by n − 1 instead of n. Dividing by n would underestimate the true variance on average.
Used by risk manager: Estimates the volatility of a position from recent returns.
Full lesson - 03beginner
Standard error of the mean
The sample mean is itself random. Its standard deviation, the standard error, is the single-observation volatility divided by the square root of the sample size.
Used by quant researcher: Puts error bars on backtest returns before trusting a signal.
Full lesson - 04beginner
Confidence interval for a mean
A range around the sample mean that, built this way on repeated samples, covers the true mean the stated share of the time. Wider means less certain.
Used by quant researcher: Reports a range for a strategy's mean return instead of a single number.
Full lesson - 05intermediate
z-test for a mean
Measure how many standard errors the sample mean is from the value the null hypothesis claims. If that distance is beyond the critical value, reject the null.
Used by quant researcher: Tests whether a signal's average return differs from zero over a long sample.
Full lesson - 06intermediate
t-statistic for a strategy's mean return
The sample mean divided by its estimated standard error. Because the volatility is estimated from the same data, it follows a t distribution rather than a normal.
Used by quant researcher: Reports the t-statistic of every backtested signal and asks for well above 2 before trading it.
Full lesson - 07intermediate
p-value
The chance of a result at least as extreme as yours if there is really nothing there. Small p-values mean luck alone rarely produces such a result.
Used by quant researcher: Screens signals by p-value, adjusting the bar for how many signals were tried.
Full lesson - 08beginner
The square-root-of-n rule
The error of an average falls with the square root of the amount of data. To halve the error you need four times the data; to cut it tenfold you need a hundred times.
Used by quant researcher: Works out how many years of backtest are needed to measure an edge of a given size.
Full lesson - 09advanced
Maximum likelihood estimation
Choose the parameter value that makes the data you observed most probable. For k winners in n trades, that is simply the win rate k/n.
Used by quant researcher: Fits distributions, volatility models such as GARCH and order-arrival rates by maximising likelihood.
Full lesson - 10intermediate
Mean squared error: bias² plus variance
The average squared error of an estimate splits into two parts: systematic error (bias, squared) and noise (variance). Accepting a little bias can remove a lot of noise.
Used by quant researcher: Shrinks noisy expected-return estimates towards zero or a prior before building portfolios.
Full lesson
Nothing matches. Try a shorter word or pick "All".
Rather not code the maths yourself? Describe your trading idea and let AI build the bot.
Build a bot with AgenKitEvery model is drawn from the formula's own code, so what you see is the exact maths. Models simplify markets on purpose. This is education, not financial advice.