Morgan Stanley Data Science Interview Questions & Answers

Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation.

Start with a sensible default such as 1e-3 for Adam, run a learning-rate range test, and use schedulers or warm-up while watching the loss curve.

Accuracy, F1 and confusion matrices for classification, BLEU or ROUGE for generation and summarisation, perplexity for language models, plus human review.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks.

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment.

Bias occurs when the sample does not represent the population. Reduce it with random or stratified sampling, higher response rates and by checking sample demographics against the population.

int, float, complex, bool, str, list, tuple, set, dict and NoneType. Mutable ones are list, set and dict; the others are immutable.

Raising the threshold improves precision but lowers recall. Fraud detection leans to recall, while spam filters often prefer precision; choose by the cost of each error.

This is overfitting. Check for data leakage, simplify the model, add regularization, get more data or augment it, use cross-validation, early stopping and feature selection.

The square root of the variance; it measures how far values typically sit from the mean, in the same units as the data.

Small anonymous one-line functions, handy as arguments to sort, map, filter or Pandas apply.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis.

The model memorises noise and fails on new data. Prevent it with more data, simpler models, regularisation, cross-validation, early stopping and dropout.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs.

Activations add non-linearity so networks can learn complex patterns. ReLU (max(0, x)) is cheap and reduces vanishing gradients compared with sigmoid/tanh; Leaky ReLU avoids 'dead' neurons.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback.

A function using yield that produces values lazily, one at a time. It saves memory when processing large files or streams because the full sequence is never held in RAM.

An optimisation method that updates weights in the opposite direction of the loss gradient. The learning rate sets the step size: too high and it diverges, too low and it is slow. Variants include SGD, mini-batch and Adam.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations.

During training it randomly switches off a fraction of neurons so the network cannot depend on any single unit, acting as a regulariser against overfitting.

Precision = TP / (TP + FP). Recall = TP / (TP + FN). Prefer recall when missing a positive is costly — for example disease screening or fraud detection.

Use the IQR rule (below Q1-1.5 IQR or above Q3+1.5 IQR), z-scores beyond 3, box plots or model-based methods such as Isolation Forest, then decide whether to cap, remove or keep them.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak.

Splitting text into units such as words, subwords or characters that are mapped to ids; subword methods like BPE handle rare words well.

Calibration checks whether predicted probabilities match real frequencies (e.g. 70% predictions are right about 70% of the time). Fix with Platt scaling or isotonic regression.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations.

Information from outside the training window, such as future data or test statistics, enters training and inflates performance. Split first, then fit scalers and encoders on training data only.

df.drop_duplicates(subset=[...], keep='first') removes repeated rows; use duplicated() first to inspect what will be dropped.

Discriminative models learn P(y given x) directly (logistic regression, SVM). Generative models learn the joint P(x, y) and can generate data (Naive Bayes, GANs).

Regularization adds a penalty on weights to reduce overfitting. L1 (Lasso) adds |w| and can push features to exactly zero. L2 (Ridge) adds w² and shrinks weights smoothly.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance.

Parametric tests assume a distribution (usually normal) and use means; non-parametric tests such as Mann-Whitney use ranks and make fewer assumptions.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation.

A range built from sample data that would contain the true parameter in a stated percentage (for example 95%) of repeated samples. It is not the probability that the parameter lies inside this one interval.

It applies Bayes' theorem assuming features are independent given the class. The assumption is rarely true but the model is fast and works well for text.

An epoch is one pass over the training data; a batch is the subset used for one update; an iteration is one weight update on a batch.

Use stratified splits, class weights, resampling (SMOTE or undersampling) and threshold tuning, and judge the model with PR-AUC or F1 instead of accuracy.

A mutex in CPython that lets only one thread execute Python bytecode at a time. Use multiprocessing for CPU-bound work and threads or asyncio for I/O-bound work.

It tries candidate splits and picks the one that most reduces impurity — Gini or entropy (information gain) for classification, variance / MSE for regression.

It normalises layer inputs per mini-batch and then rescales them, stabilising and speeding up training and allowing higher learning rates.

Gradients shrink as they pass through many layers (for example with sigmoid), so early layers barely learn. ReLU, residual connections and good initialisation help.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity.

It checks whether observed category counts differ from expected counts, used for independence between two categorical variables and for goodness of fit.

A set stores unique values with no keys; a dictionary stores key-value pairs with unique keys. Both use hashing, so lookups are O(1) on average.

Bagging trains models in parallel on bootstrap samples and averages them (reduces variance, e.g. Random Forest). Boosting trains models one after another, each fixing the previous errors (reduces bias, e.g. XGBoost).

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context.

P(A given B) = P(B given A) x P(A) / P(B). In a medical test it converts test accuracy and disease prevalence into the real chance a positive result means disease.

Type I = false positive (rejecting a true null), probability α. Type II = false negative (missing a real effect), probability β. Power = 1 − β.

It computes the gradient of the loss with respect to every weight by applying the chain rule backwards through the network, so gradient descent can update them.

pd.pivot_table(df, index='region', columns='month', values='sales', aggfunc='sum') reshapes long data into a summary matrix; melt() does the reverse.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix.

Correlation measures linear association between variables (-1 to 1). A third factor, reverse causality or chance can create it, so causation needs controlled experiments.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns.

Use target or frequency encoding with cross-validation, hashing, embeddings or grouping rare categories, instead of huge one-hot matrices.