how to positive definite array in Vector autoregression?

Viewed 19
from statsmodels.tsa.vector_ar.var_model import VAR
max_lag = 20
# occ_data is a dataframe by the size of 52584 rows × 5 columns
var_model = VAR(occ_data)
# select the best lag order
lag_results = var_model.select_order(max_lag)
selected_lag = lag_results.aic
print(selected_lag) 

After running this I get the error: 5-th leading minor of the array is not positive definite

for this line: lag_results = var_model.select_order(max_lag)

How do I make the array to positive definite? Thanks for the help!

occ_data:

    BGT North of NE 70th Total  Ped South   Ped North   Bike North  Bike South
Date                    
2014-01-01 00:00:00 0.002665    0.000243    0.000186    -1.901329e-05   0.002255
2014-01-01 01:00:00 -5.997088   1.000224    -1.999776   -1.000000e+00   -3.997536
2014-01-01 02:00:00 0.002912    -0.999776   0.000224    -1.000000e+00   2.002464
2014-01-01 03:00:00 10.002912   0.000224    0.000224    4.260205e-09    10.002464
2014-01-01 04:00:00 0.002912    0.000224    0.000224    4.260205e-09    0.002464
... ... ... ... ... ...
2019-12-31 19:00:00 -7.997536   -2.000000   -4.999104   1.000896e+00    -1.999328
2019-12-31 20:00:00 -3.997536   -2.000000   -0.999104   -9.991039e-01   0.000672
2019-12-31 21:00:00 1.002464    1.000000    2.000896    8.960573e-04    -1.999328
2019-12-31 22:00:00 -1.997536   -1.000000   -1.999104   8.960573e-04    1.000672
2019-12-31 23:00:00 1.002464    1.000000    0.000896    1.000896e+00    -0.999328

VAR: which stands for Vector Auto-Regression - is a statistical model used to capture the relationship between multiple quantities as they change over time. VAR is a type of stochastic process model. VAR models generalize the single-variable (univariate) autoregressive model by allowing for multivariate time series. VAR models are often used in economics and the natural sciences.

0 Answers
Related