gavindiaz/QuanTAlib
1
0
Fork 0
mirror of https://github.com/mihakralj/QuanTAlib.git synced 2026-08-23 21:18:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: resolve build and test errors

Browse Source 
- Sar.Quantower.Tests.cs: add missing opening quote on string literal (line 48)
- Exports.cs: rename Correlation.Batch → Correl.Batch (CS0103)
- Ad.Validation.Tests.cs: fix Ooples OutputValues key "Ad" → "Adl"
This commit is contained in:
Miha Kralj
2026-03-16 12:45:13 -07:00
parent 3b0cdca567
commit 6f0a339c9b
131 changed files with 1570 additions and 1571 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/statistics/adf/Adf.md
+3 -1
View File
@@ -13,7 +13,9 @@
| **PineScript** | [adf.pine](adf.pine) |
- Tests the null hypothesis that a time series contains a unit root (non-stationary). Output near **0** → stationary; output near **1** → unit root.
- **Similar:** [Hurst](../hurst/Hurst.md), [Cointegration](../cointegration/Cointegration.md) | **Complementary:** Z-Score, Variance | **Trading note:** ADF < 0.05 confirms mean-reversion suitability.
- **Similar indicators:** [Hurst](../hurst/Hurst.md), [Cointegration](../cointegration/Cointegration.md)
- **Complementary indicators:** [Z-Score](../zscore/Zscore.md), [Variance](../variance/Variance.md)
- **Trading note:** ADF < 0.05 confirms mean-reversion suitability.
- Validated against Python `statsmodels.tsa.stattools.adfuller` reference implementation.
The Augmented Dickey-Fuller test is the gold standard for detecting whether a financial time series is stationary or contains a unit root. Unlike the original Dickey-Fuller test, the augmented version includes lagged difference terms $\Delta y_{t-i}$ to absorb serial correlation, ensuring the test statistic follows the correct distribution. The p-value output uses MacKinnon (1994, 2010) polynomial interpolation with a standard normal CDF approximation, providing machine-precision results without lookup tables.