M$: It’s not a bug, it’s a feature! [Software]

posted by Helmut Homepage – Vienna, Austria, 2019-12-17 15:00 2a02:8388:6bc2:ce80:6c4e:e4df:42f4:6510 – Posting: # 20985
Views: 539

Hi all,

known for ages. All (‼) versions of Excel round the degrees of freedom down to the nearest integer.

df <- seq(10, 11, 0.2)
q  <- 4.5
res <- data.frame(df = df, correct = pt(q, df), Excel = pt(q, floor(df)))
print(res, row.names = FALSE)

[image]  df   correct     Excel
10.0 0.9994284 0.9994284
10.2 0.9994561 0.9994284
10.4 0.9994818 0.9994284
10.6 0.9995059 0.9994284
10.8 0.9995284 0.9994284
11.0 0.9995494 0.9995494

Quoting Martin:

Never never never never use Excel. Not even for calculation of arithmetic means.


Congratulations to M$ for making a continuous function discrete. :thumb down:

I had fun comparing papers by regulators based on Luther Gwaza’s Excel-Sheet for adjusted indirect comparisons with my R-package. Satterthwaite’s degrees of freedom are practically never* integers. Remember this goody of the EMA?

Results obtained by alternative, validated statistical programs are also acceptable except spreadsheets because outputs of spreadsheets are not suitable for secondary assessment.

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

 Admin contact
20,333 posts in 4,270 threads, 1,401 registered users;
online 4 (0 registered, 4 guests [including 3 identified bots]).
Forum time (Europe/Vienna): 03:56 CET

No matter what side of the argument you are on,
you always find people on your side
that you wish were on the other.    Thomas Berger

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz