How deeply does Causal Inference for Data Science cover k-nearest neighbors?

Explains separate-group models, counterfactual predictions, averaging, and cross-validated k selection; stops short of broader algorithmic internals.

K-nearest neighbors

As explained in Causal Inference for Data Science

K-nearest neighbors is a prediction method that uses the outcomes of the closest training observations to predict an input's outcome.

K-nearest neighbors (kNN) predicts an outcome from the outcomes of the k most similar observations. In causal matching, similarity is measured using propensity scores. section 5.3.4

A causal matching workflow uses kNN to improve on single-patient matches:

Separate observations into treated and control datasets. section 5.3.4
Train outcome-prediction models on each group, using propensity score as the feature and outcome O as the target. section 5.3.4
Choose k as a hyperparameter, using cross-validation rather than relying on one potentially extreme match. section 5.3.4
Predict counterfactual outcomes across groups, calculate individual treatment effects, and average them to estimate the ATE. section 5.3.4

Treatment-effect matching

For a treated patient with propensity score sᵢ, use the kNN trained on controls to predict the expected untreated outcome at sᵢ. Reverse the process for control patients, then average the resulting treatment effects. section 5.3.4

Table 5.4 Treatment data to be used with kNN, D_t_·_{n_t} is the number of patients in the treatment group.

O	S
o ₁	s ₁
…	…
O _{n_t}	s _{n_t}

Inspect the treatment dataset containing outcomes and propensity scores used to train one kNN model. section 5.3.4

Table 5.5 Control data to be used with kNN, D_c_·_{n_c} is the number of patients in the control group.

O	S
o ₁	s ₁
…	…
O _{n_c}	s _{n_c}

Inspect the corresponding control dataset used to train the comparison kNN model. section 5.3.4

How Causal Inference for Data Science covers it

Cover of Causal Inference for Data Science

liveBook manning.com

Read Causal Inference for Data Science on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Outlier Detection in PythonA k-nearest-neighbors algorithm scores a record using its average distance to its k nearest neighboring records.
liveBook manning.com

Building LLM Applications with DSPyThe k examples most similar to a query according to a measure of similarity.

Compare how 3 books explain k-nearest neighbors →

Related concepts

KNN optimizerused by · A DSPy optimizer that selects demonstrations dynamically by retrieving the k training examples most similar to each language-model call's inputs.
radius detectoralternative to · A radius detector identifies or scores observations according to the number or arrangement of neighboring observations within a specified distance radius.
propensity score methodimplemented by · A propensity-score method is a causal-inference technique that estimates treatment effects by matching, weighting, stratifying, or otherwise adjusting units according to their observed treatment probabilities.
parameterconfigured by · A model parameter is a numerical value, such as a weight or bias, that is learned or set to determine how a model computes outputs.
lazy evaluationuses · Lazy evaluation is an evaluation strategy that defers computing an expression or value until its result is needed, potentially avoiding the computation entirely.

This explainer was generated by Manning's knowledge graph from the text of Causal Inference for Data Science. The section links point to the parts of the book it draws on.