Questions tagged [winsorizing]
Winsorizing is a kind of data transformation used in robust/resistant statistics. Extreme values in the sample is replaced by some chosen data quantile(s). See https://en.wikipedia.org/wiki/Winsorizing
19 questions with no upvoted or accepted answers
4
votes
0
answers
2k
views
Treating outliers for time series forecasting in Python
What is the best way to treat outliers in a time series forecasting model? In particular, for product demand modeling?
Based on what I've read so far, the following methods can be applied:
...
- 749
2
votes
0
answers
2k
views
Dealing with outliers: Interquartile range normalization vs. Winsorization
According to this page -- "When a data set has outliers, variability is often summarized by a statistic called the interquartile range, which is the difference between the first and third ...
- 841
2
votes
0
answers
424
views
Does pre winsorising of a variable help for a logistic regression?
I am wondering if winsorising makes a difference in a logistic regression.
In a situation where I am looking at the individual contribution, looking at their individual discriminatory power (...
- 210
2
votes
0
answers
157
views
Optimizing Robust Statistics
A robust paired t-test is a better choice for skewed distributions than the conventional paired t-test (e.g Fradrette, Keselman, Lix, & Wilcox, 2003). One version of the robust test uses a trimmed ...
- 640
2
votes
0
answers
551
views
Transformation and/or Winsorizing?
I want to compare two group of 24 and 28 people with t-test on type of activity (5 different's types of activity and a total), later one the same value will be use in regression logistic.
If you ...
- 41
2
votes
0
answers
211
views
Robust Estimators - Winsorized Variance degree of freedom (df)
this is my first question on this site.
So, I'm currently working on my final year thesis, and it was on Robust statistics. In my work, I will use Trimmed Mean, Winsorized Mean and Winsorized ...
- 21
1
vote
0
answers
157
views
Valid approach: Winsorizing data for main analysis and then doing sensitivity analysis without winsorizing?
I've got a variable with psychological data (N=75) which is distributed pretty symmetrical, but has very few cases with very extreme values, more extreme to the left tail. But nevertheless this data ...
- 11
1
vote
1
answer
163
views
Removing outliers in several groups and for several features
I'm unsure on how to remove or winsorize outliers. Let's say I have 2 groups, treated and control. And I measure feature1 and feature2 for both.
How should I handle outliers? For each group and each ...
- 475
1
vote
0
answers
482
views
How to optimally choose winsorization thresholds for different metrics in large scale A/B testing platform
I work on our A/B testing platform where we have implemented one-sided winsorization broadly across all continuous variables (capped at 95th percentile). While that's a common cut-off, some of our ...
- 11
1
vote
0
answers
120
views
Name for the opposite of Winsorizing?
For some regressions we find it useful to focus on extreme values, and so we discard middling dependent values (which we might call "noise") from data in order to find relationships that hold at data ...
- 1,176
1
vote
0
answers
323
views
Winsorization when we run regressions by size group
I have a sample that consists of large, medium, and small firms and i want to run a separate regression for each size group. When I winsorize a variable should I do it for the whole sample (i.e. ...
- 43
1
vote
0
answers
110
views
Winsorizing, just the outlier or all the value?
I have an outlier in my data set. I want to use the winsorizing quartile (to change the outlier to the 5th% and/or 95th%).
Looking at the quartiles, sometimes I have more values than just my ...
- 41
0
votes
0
answers
42
views
Winsorizing outliers across multiple analyses: once or multiple times? (SPSS)
I have a 2×2 experimental design with four conditions and eight outcome variables. I’m supposed to winsorize outliers, but I’m confused about how many times this needs to be done because I’m ...
- 21
0
votes
1
answer
237
views
Can I apply both winsorization and CUPED to my experiment results?
Our current experimentation platform currently has winsorization implemented to reduce "whale effects" on metrics like revenue and volume. We are also interested in applying CUPED to further ...
- 11
0
votes
0
answers
1k
views
Winsorizing or taking the logarithm first?
I testing if I can describe the StockPRice with EPS (=earnings per share), BookValuePS an ESGscore.
Before I start I winsorized all my variables. Now I want to take the loagrithm of e.g. BookValuePS ...
0
votes
0
answers
701
views
How to choose cut off for winsorization/ flooring- capping? What is the impact of variable distribution on the decision
To perform logistic regression I wish to winsorize outliers in independent/ explanatory variables by flooring and capping independent variables.
Can you suggest how I should choose cut-off for ...
0
votes
0
answers
901
views
Winsorizing data in small sample
I have a relatively small sample of panel data (quarterly data for 68 firms over 7 years). My dependent variable is positively skewed. In order to limit the influence of observations with large values,...
- 497
0
votes
0
answers
722
views
Winsorization to remove spikes in time series
In product demand forecasting, is it valid to use winsorization to remove large outliers (spikes) in the data? I understand that the spikes may be due to holiday effects (e.g. people will buy more ...
- 749
0
votes
0
answers
549
views
Using trimmed means and Winsorized variances to compute standardisation of data
I am looking into the pros and cons of each normalisation technique for work and it got me thinking. What if I used trimmed means and the sqrt of Winsorized variances to compute the standardised data? ...
- 317