espin086 / espin086/Experiments

Create One Sided T Test

Open
#2 0 comments 0 reactions 1 assignee Claimed by @espin086 View on GitHub
Dominant language
Python
Stars
0
Forks
0
PR merge metrics
No merged PRs in 30d

Description


One tailed test: Tests for better treatment performance (use with discretion - only tests one direction) |   |   |   |   |   |  
-- | -- | -- | -- | -- | -- | --
  | Enter Data |   |   |   |  
  | Test | Control |   |   |   |  
Test Results: | 0.51 | 0.5 |   |   |   |  
Confidence Level: | 0.9 |   |   |   |   |  
Sample Size: | 100000 | 100000 |   |   |   |  
  |   |   | p-value= | =IF((1-NORMSDIST(E47))<0.0001,"<0.0001",(1-NORMSDIST(E47))) |   |  
Answer: | =VLOOKUP(E51,E49:F50,2) |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
Should I use a one-tailed test? |   |  
  |   |   |   |   |   |  
Use a one-tailed test if you are interested in an effect in only one direction and not the other, even if you see a large impact in the other direction. Examples:1. Does $25M marketing spend improve product acquisition rate?2. A new tool has been developed that is cheaper than the existing tool. Is it less effective than the existing tool?Sources:https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/j.2041-210X.2010.00014.xhttps://stats.idre.ucla.edu/other/mult-pkg/faq/general/faq-what-are-the-differences-between-one-tailed-and-two-tailed-tests/*Note that switching to a one-tailed test for the sole purpose of making an otherwise nonsignificant result significant is not appropriate. |   |  
  |  
  |  
  |  
  |  
  |  
  |  
  |  
  |  
  |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
Work Area
  |   |   |   |   |   |  
Test Proportion | P | =+D9 |   |   |   |  
Control Proportion | T | =+E9 |   |   |   |  
Difference |   | =+E37-E38 |   |   |   |  
Critical Point | a = | =ABS(NORMSINV(1-D10)) |   |   |   |  
Sample size control | N1 | =+D11 |   |   |   |  
Sample Size Test | N2 | =+E11 |   |   |   |  
Pooled Sample Proportion | p* | =(+E41*D9+E42*E9)/(E41+E42) |   |   |   |  
  |   |   |   |   |   |  
Pooled Standard Error | SE(ofP*) | =SQRT((E43*(1-E43)*(E41+E42))/(E41*E42)) |   |   |   |  
  |   |   |   |   |   |  
Test Statistic | (P-T)/SE(of p*) | =(E37-E38)/E45 |   |   |   |  
  |   |   |   |   |   |  
  | TABLE | 0 | NOT SIGNIFICANT |   |  
  |   | 1 | SIGNIFICANT |   |   |  
  | ANSWER | =IF(E47>=E40,1,0) |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  
  |   |   |   |   |   |  

 


USAA Classification: Internal

Disclaimer: This email and any attachments are the property of USAA and may contain confidential and/or privileged material. If you are not the intended recipient, any use, disclosure or copying of this email or any attachments is unauthorized. If you received this email in error, please immediately notify the sender and delete the email and any attachments from your computer.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.