Appendix B — NumPy arrays

This appendix provides a quick-reference guide to the main NumPy array tools used in this book. We focus on functions, methods, and attributes used directly with NumPy arrays or used to create arrays for numerical work. To keep the list compact, we omit random-number-generator methods (for example, rng.normal()) and constants or dtypes such as np.nan or np.int8.

Table B.1: Common NumPy functions used with arrays

Function	Purpose	Where used
`np.array()`	Build an array from a sequence	Data fundamentals Statistical inference Instrumental variables Extreme values and sensitivity analysis
`np.arange()`	Integer sequence	Data fundamentals Accruals Earnings management Instrumental variables Natural experiments revisited Matching Extreme values and sensitivity analysis
`np.asarray()`	Convert to an array	Prediction Instrumental variables Extreme values and sensitivity analysis
`np.empty()`	Uninitialized array	Accruals Natural experiments revisited
`np.full()`	Fill with one value	Accruals Earnings management
`np.zeros()`	Zeros array	Extreme values and sensitivity analysis
`np.ones()`	Ones array	Regression fundamentals Statistical inference Accruals Earnings management Natural experiments revisited Matching
`np.repeat()`	Repeat values	Accruals Instrumental variables Natural experiments revisited
`np.tile()`	Repeat blocks	Natural experiments revisited
`np.linspace()`	Evenly spaced values	Statistical inference Natural experiments revisited
`np.column_stack()`	Bind columns into a matrix	Regression fundamentals Statistical inference Accruals Earnings management Instrumental variables Natural experiments revisited Matching
`np.c_[]`	Column-bind using index syntax	Natural experiments revisited
`np.sum()`	Sum values	Data fundamentals Statistical inference
`np.mean()`	Mean	Statistical inference
`np.std()`	Standard deviation	Statistical inference
`np.median()`	Median	Data fundamentals
`np.quantile()`	Quantile or cutoff	Causal inference
`np.cov()`	Covariance matrix	Data fundamentals Extreme values and sensitivity analysis
`np.sqrt()`	Square root	Data fundamentals Statistical inference Earnings management Extreme values and sensitivity analysis
`np.exp()`	Exponential	Beyond OLS
`np.abs()`	Absolute value	Earnings management Statistical inference Extreme values and sensitivity analysis
`np.where()`	Conditional choice	Statistical inference
`np.minimum()`	Elementwise minimum	Instrumental variables
`np.argmax()`	Index of maximum	Instrumental variables
`np.diag()`	Diagonal of a matrix	Statistical inference Extreme values and sensitivity analysis
`np.outer()`	Outer product	Extreme values and sensitivity analysis
`np.fill_diagonal()`	Fill matrix diagonal	Extreme values and sensitivity analysis

Table B.2: Common NumPy array methods

Method	Purpose	Where used
`.sum()`	Sum values	Data fundamentals
`.mean()`	Mean	Data fundamentals Beyond OLS
`.max()`	Maximum	Data fundamentals
`.min()`	Minimum	Data fundamentals
`.astype()`	Convert dtype	Beyond OLS Prediction
`.item()`	Extract a scalar	Instrumental variables

Table B.3: Common NumPy array attributes

Attribute	Purpose	Where used
`.dtype`	Stored data type	Data fundamentals
`.size`	Number of elements	Data fundamentals
`.shape`	Array dimensions	Data fundamentals Regression fundamentals Instrumental variables Natural experiments revisited
`.T`	Matrix transpose	Regression fundamentals Statistical inference