Skip to content

Latest commit

 

History

History
23 lines (18 loc) · 1.22 KB

README.md

File metadata and controls

23 lines (18 loc) · 1.22 KB

Excel LAMBDA functions for timeseries analysis

This repo contains three functions that might help you conduct timeseries analysis in Excel. Specifically:

  • ACFX: Compute the autocorrelation function on the residuals of an OLS regression
  • PACFX: Compute the partial autocorrelation function on the residuals of an OLS regression
  • CHOLESKY: Compute the Cholesky decomposition of a square, symmetric, positive definite matrix

For context on how to interpret ACF and PACF see https://spureconomics.com/interpreting-acf-and-pacf-plots/.

The definitions found in the .lambda are not suitable to copy directly into an Excel spreadsheet because they contain comments, newlines and exceess spaces. The Python script make_lambda.py strips all spaces, newlines and anything on a line appearing after // from the file, making it suitable to copy-and-paste into an Excel named range. The useage is:

python make_lambda.py < WIBBLE.lambda > WIBBLE.min.lambda

Warning: do not use make_lambda.py on functions that manipulate strings, especially ones that contain string literals with spaces in them since it will mangle them. All the functions in this repo are safe, however.