Data science cheat sheet (Business data science)

1. 1Code for data: Python or RThe working set you touch every day: notebooks, arrays, data frames, filtering, grouping, joining and reshaping, plus an environment someone else can rebuild.9 concepts

   1. 1Notebook workflowCode, output, charts and notes in one document you run cell by cell: Jupyter or Colab for Python, RStudio with Quarto or R Markdown for R.
   2. 2Arrays and vectorsA block of values of one type: a NumPy array in Python, an atomic vector in R.
   3. 3Vectorised operationsWriting one expression over a whole column instead of looping over rows, so the work runs in fast compiled code.
   4. 4DataFrame and tibbleA table in memory: named columns, each with one type, and rows that line up across them. pandas in Python, tibbles in R.
   5. 5Selecting and filtering rowsKeeping the columns you need and the rows that meet a condition: loc and boolean masks in pandas, select and filter in dplyr.
   6. 6Group by and aggregateSplitting rows into groups by one or more columns, computing a summary for each group and combining the results into a new table.
   7. 7Merging tablesCombining two data frames on a key column, like a SQL join: merge in pandas, left_join and friends in dplyr.
   8. 8Long and wide formatWide data has one column per period or category; long data has one row per observation with a column saying which one.
   9. 9Reproducible environmentA per-project record of the exact package versions an analysis used, so it runs the same on another machine or next year.

   9 in this chapterOpen this chapter
2. 2Statistics and probability you useSummaries, distributions, uncertainty and tests: the part of statistics that decides whether a pattern in your data is real, and which test to reach for.10 concepts

   1. 1Mean, median, standard deviation and IQRThe mean is the sum over n; the median is the middle value; the standard deviation is the typical distance from the mean; the IQR is Q3 minus Q1, the spread of…
   2. 2Common distributionsShapes that describe how a random quantity behaves: normal for sums of many small effects, binomial for counts of successes in n tries, Poisson for counts of…
   3. 3Conditional probability and Bayes' ruleP(A|B) is the probability of A given that B happened.
   4. 4Sampling distribution and the CLTIf you repeated your sample many times, the sample mean would vary; that spread is its sampling distribution.
   5. 5Standard errorThe standard deviation of an estimate across repeated samples.
   6. 6Confidence intervalA range built so that 95% of intervals made this way would contain the true value.
   7. 7Hypothesis test and p-valueAssume no effect (the null hypothesis) and ask how surprising your data would be.
   8. 8Type I, type II errors and powerA type I error is finding an effect that is not there (rate alpha, usually 5%).
   9. 9Which test whenTwo group means: Welch t-test.
   10. 10Correlation versus causationCorrelation measures how two variables move together, from -1 to 1.

   10 in this chapterOpen this chapter
3. SQL for data workPulling the data you need from a warehouse or database: aggregates, joins at the right grain, window functions, and running SQL from your notebook.6 conceptsOpens with credits
4. Data cleaning and EDAGetting raw data into a shape you can trust, then exploring it before you model anything: missing values, duplicates, types, outliers and first looks.7 conceptsOpens with credits
5. Visualisation and storytellingChoosing the chart that answers the question, drawing it with your plotting library, showing uncertainty, and writing up the finding so someone can act on it.6 conceptsOpens with credits
6. Machine learning with scikit-learn or tidymodelsThe core modelling loop for tabular data: split, baseline, fit, validate, compare. Enough to model well; the machine learning roadmap goes deeper.8 conceptsOpens with credits
7. Experiments and A/B testingDesigning a fair experiment, sizing it, running it without peeking, checking it is healthy and reading the result honestly.9 conceptsOpens with credits
8. Features and model evaluationTurning raw columns into useful inputs without leaking the answer, and choosing the metric that matches the decision.9 conceptsOpens with credits
9. Working with messy real dataGetting data from APIs, open portals and the web responsibly, documenting where it came from, and knowing the biases it carries.5 conceptsOpens with credits
10. Portfolio and first roleProjects that show judgement, write-ups people finish reading, a Kaggle profile used well, and the routes into a first data science job.7 conceptsOpens with credits
11. 11Glossary, A to ZEvery concept in the sheet and the smaller terms it uses, in one list.132 terms

Statistics, pandas, ML, experiments.