1. In the first code cell,

outcome_controls =115/num_controls

should be outcome_controlled to match later code.

2. In the first code cell,

std_treatment_effect = np.sqrt(var_treatment_effect

is missing the right parenthesis.

3. In the first code cell,

print(f"95 % confidence interval is [{CI_delta[0])}, {CI_delta[1]}]")

has an extra parenthesis after CI_delta[0].

4. In the Pfizer/BNTX Covid-19 RCT markdown cell, "averal" is written instead of "overall."

5. In the cell below, "vaccie" is written instead of "vaccine."

For the polishing of the notebooks, a list of the notebooks is here:

https://docs.google.com/spreadsheets/d/1RL01oaOYDZ2PM9CtjOShPM_dOrxbFbnd72dHPMOpPds/edit?usp=sharing
and the notebooks sections in the book (might slightly differ).

Tasks:

1. Checking all notebooks (chapter by chapter)
   a) if they run without error and are correct, in particular

• compare results with the results in the book
• compare results between R and Python versions
• for long computations in the notebooks raise note
• sufficient documentation of the notebooks
  b) they should all be stored at github
  https://github.com/CausalAIBook/MetricsMLNotebooks
  c) For b) translation of kaggle scripts (R) to Jupyter notebooks for use colab
  script: https://github.com/CausalAIBook/MetricsMLNotebooks/blob/main/Rmd2ipynb
  d) Updating the entries and links in the google table above; adding missing notebooks

2. If missing, translation of notebooks from R to Python or the other way around, so that R and Python scripts are available (but might differ)
3. Preparation of new notebooks (Oliver is working on DiD; RD next)

In this issue, we will update the notebooks housed on kaggle for PM1 following the updated .Rmd files in PM1.

@MartinSpindler Can you assign a task to the relevant person to take the newly updated .Rmd files in directory PM1 and convert them to Kaggle notebooks? We will then want to change the Python code and notebooks to identical structure.

The current code links to github's wage data which is only an RData dataset, then the code reads:

Data Analysis
Set the following file_directory to a place where you downloaded https://github.com/CausalAIBook/MetricsMLNotebooks/blob/main/PM1/wage2015_subsample_inference.rdata

file_dir = None
df = pd.read_csv(file_dir)

I found two solutions:

1. Import RData to Pandas ready package (taken from my collab code):

!pip install pyreadr
import pyreadr
import numpy as np
import pandas as pd
temp = pyreadr.read_r('/content/wage2015_subsample_inference.Rdata')
df = temp['data']

2. Upload a CSV

In the coding cell of "# Print the R^2", there are two additional right parenthesizes, which are in line 5 and line 22.

In cell 2 of pm2-notebook-jannis.ipynb

growth_read = read_r(wget("https://github.com/CausalAIBook/MetricsMLNotebooks/blob/main/data/GrowthData.rda?raw=true"))

should be

growth_read = read_r(wget.download("https://github.com/CausalAIBook/MetricsMLNotebooks/blob/main/data/GrowthData.rda?raw=true"))