Skip to content
This repository was archived by the owner on Mar 5, 2024. It is now read-only.

Use ddof=1 in np.var for the unbiased estimator of the variance #30

Open
coldfire79 wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Use ddof=1 in np.var for the unbiased estimator of the variance #30

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dc0219a
add in np.var to get the unbiased estimator of the variance as in th…
coldfire79 Sep 19, 2022
9cabb30
revert the test file
coldfire79 Sep 19, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions combat/pycombat.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,7 +88,7 @@ def compute_prior(prior, gamma_hat, mean_only):
if mean_only:
return 1
m = np.mean(gamma_hat)
s2 = np.var(gamma_hat)
s2 = np.var(gamma_hat, ddof=1)
if prior == 'a':
return (2*s2+m*m)/s2
elif prior == 'b':
Expand Down Expand Up @@ -521,11 +521,11 @@ def fit_model(design, n_batch, s_data, batches, mean_only, par_prior, precision,
else:
for i in batches: # feed incrementally delta_hat
list_map = np.transpose(np.transpose(s_data)[i]).var(
axis=1) # variance for each row
axis=1, ddof=1) # variance for each row
delta_hat.append(np.squeeze(np.asarray(list_map)))

gamma_bar = list(map(np.mean, gamma_hat)) # vector of means for gamma_hat
t2 = list(map(np.var, gamma_hat)) # vector of variances for gamma_hat
t2 = gamma_hat.var(axis=1, ddof=1).flatten().tolist()[0] # vector of variances for gamma_hat

# calculates hyper priors for gamma (additive batch effect)
a_prior = list(
Expand Down