Add summary statistics functions

ONEcampaign / toughest_places_index

New analysis from the ONE Campaign presents the toughest places to feed a family

MIT License

1 stars 0 forks source link

Add summary statistics functions #26

Closed jm-rivera closed 2 years ago

jm-rivera commented 2 years ago

Functions which allow us to quickly print/get useful information on the data for a specific indicator and/or dimension of indicators.

For example:

How much data is missing (overall and maybe broken-down by region and income level)
Are there any outliers
Are indicators in a dimension highly correlated
Is the variance in a dimension highly explained by a minority of indicators

@lpicci96 anything else we could be looking at?

lpicci96 commented 2 years ago

how many 0s (if any)
Plotting functions? box plot + histogram

jm-rivera commented 2 years ago

Thanks @lpicci96. Do you want to tackle this one? Let's park plotting for now unless it is plotting that reveals something about the data that we can't get from the other summaries.

lpicci96 commented 2 years ago

@jm-rivera preferred method to determine outliers? Or we can add a "method" parameter to specify the calculation - IQR, z score etc

jm-rivera commented 2 years ago

@jm-rivera preferred method to determine outliers? Or we can add a "method" parameter to specify the calculation - IQR, z score etc

Would prefer the 'method' option that way we can iterate with the same callable