← Bank of America Interview Insights

Bank of America·Data Scientist·Technical Phone Screen·Intermediate

Intermediate
Jun 2026

Summary

Went through what seemed like a theory-heavy screening for a Data Scientist role at Bank of America. Lots of stats and ML fundamentals, the kind of stuff you think you know cold until someone asks you to write it out precisely.

Questions Asked (9)

Q1

You roll two fair six-sided dice. What is the probability that the two dice show different values? And what is the probability that one specific die shows a strictly higher value than the other?

Product Analytics & Metrics
Author's notes

The first part is easy enough, 30 out of 36 outcomes are unequal so 5/6.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the sample space of 36 equally likely outcomes for two fair six-sided dice. For the first question, compute the probability of different values by subtracting the probability of matching values from 1. For the second question, recognize that by symmetry, the probability that one specific die is strictly higher than the other is half the probability of different values.

Pro tip: Explicitly state the assumption of independence and fairness, and mention that the symmetry argument relies on the dice being identical and fair. This shows rigor and avoids ambiguity.

1. Define the sample space

State that there are 36 equally likely outcomes when rolling two fair six-sided dice, as each die has 6 independent outcomes.

2. Compute probability of different values

Calculate the number of outcomes where the dice show different values: 6 * 5 = 30. Then divide by 36 to get 5/6.

3. Compute probability of one die strictly higher

Use symmetry: since the dice are fair and identical, the probability that die A > die B equals the probability that die B > die A. These two events are disjoint and together make up the 'different values' event. Thus, each has probability (5/6)/2 = 5/12.

4. Verify with enumeration (optional)

Optionally, list or count the outcomes where one die is strictly higher (e.g., 15 outcomes for each direction) to confirm the result.

Key Points to Mention

  • Sample space of 36 equally likely outcomes
  • Independence and fairness of dice
  • Complement rule for 'different values' (1 - P(same))
  • Symmetry argument for 'one die strictly higher'
  • Probability of same value is 1/6 (6 outcomes out of 36)
  • Final probabilities: 5/6 for different values, 5/12 for one die strictly higher

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

Define mean and variance of a random variable.

Product Analytics & Metrics
Author's notes

Felt like a warmup but they wanted proper definitions, not hand-waving.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start with clear, concise definitions of mean and variance, using both mathematical formulas and intuitive explanations. Then, connect these concepts to practical applications in data science, especially in the context of banking and finance. Finally, briefly mention related concepts like standard deviation and expected value to demonstrate depth of understanding.

Pro tip: Emphasize that variance is in squared units, so standard deviation is often more interpretable; this shows attention to practical detail. Also, relate the concepts to real-world banking scenarios, such as risk assessment or customer lifetime value, to stand out.

1. Define Mean

State that the mean (expected value) is the weighted average of all possible values of a random variable, where weights are probabilities. Provide the formula: E[X] = Σ x * P(X=x) for discrete, or ∫ x f(x) dx for continuous.

2. Define Variance

Define variance as the expected squared deviation from the mean, measuring spread. Provide the formula: Var(X) = E[(X - μ)^2] = E[X^2] - (E[X])^2.

3. Explain Intuition

Describe mean as the 'center of mass' or long-run average, and variance as the average squared distance from that center, indicating how spread out the values are.

4. Connect to Data Science

Discuss how mean and variance are used in data science, e.g., in feature scaling, anomaly detection, portfolio risk (variance as risk), and model evaluation.

5. Mention Related Concepts

Briefly note standard deviation (square root of variance) for interpretability, and possibly covariance for multivariate cases.

Key Points to Mention

  • Mean is the expected value, a measure of central tendency.
  • Variance is the expected squared deviation from the mean, a measure of dispersion.
  • Formulas: E[X] = Σ x P(X=x) or ∫ x f(x) dx; Var(X) = E[(X-μ)^2] = E[X^2] - μ^2.
  • Variance is in squared units, so standard deviation is often used for interpretability.
  • In finance, variance is often used as a measure of risk or volatility.
  • Mean and variance are foundational for many statistical methods and machine learning algorithms.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

Is the standard sample variance estimator biased, and if so what correction makes it unbiased?

Product Analytics & Metrics
Author's notes

Yes, dividing by n gives a biased estimator because it underestimates the true variance.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by directly answering 'yes' and explaining that the standard sample variance with denominator n is biased because it underestimates the true population variance. Then introduce the unbiased estimator using denominator n-1 (Bessel's correction) and briefly explain why it works, perhaps with a simple example or intuition.

Pro tip: Mention that while n-1 is unbiased for variance, the sample standard deviation (square root of unbiased variance) is still biased for the population standard deviation due to Jensen's inequality. This shows depth and awareness of subtle statistical issues.

1. Define the estimators

Clearly state the standard sample variance formula with denominator n and the corrected version with denominator n-1. Use proper notation: s_n^2 = (1/n)Σ(X_i - X̄)^2 and s_{n-1}^2 = (1/(n-1))Σ(X_i - X̄)^2.

2. Explain the bias

Show that E[s_n^2] = σ^2 - σ^2/n = ((n-1)/n)σ^2, so it underestimates the true variance. This demonstrates that the bias is negative and depends on sample size.

3. Introduce the correction

State that using n-1 in the denominator (Bessel's correction) yields an unbiased estimator: E[s_{n-1}^2] = σ^2. Explain that this correction accounts for the fact that the sample mean is estimated from the same data, reducing the degrees of freedom by one.

4. Discuss implications and limitations

Mention that while s_{n-1}^2 is unbiased for variance, its square root is not unbiased for standard deviation. Also note that for large n, the bias of s_n^2 becomes negligible, but for small samples it matters.

Key Points to Mention

  • Bessel's correction (n-1 denominator)
  • Degrees of freedom
  • Expected value calculation: E[s_n^2] = ((n-1)/n)σ^2
  • Unbiasedness of s_{n-1}^2 for σ^2
  • Bias of sample standard deviation as an estimator of population standard deviation
  • Asymptotic unbiasedness as n → ∞

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

If X and Y are each normally distributed and their correlation is zero, does that mean they are independent? Under what condition does zero correlation actually imply independence?

Product Analytics & MetricsTechnical Trade-offs
Author's notes

No, zero correlation does not generally imply independence.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying that zero correlation does not generally imply independence, then explain the special case of jointly normal variables where it does. Use a counterexample to illustrate the general case and state the condition precisely.

Pro tip: Mention that in finance, assuming independence from zero correlation can be dangerous, especially during market stress when correlations can spike, so always check for joint normality or use copulas.

1. Define the concepts

Briefly define correlation and independence, emphasizing that correlation measures linear dependence while independence means no relationship whatsoever.

2. Answer the general case

State that zero correlation does not imply independence in general, and provide a counterexample such as Y = X^2 with symmetric X.

3. Introduce the special case

Explain that if X and Y are jointly normally distributed, then zero correlation does imply independence.

4. State the condition clearly

Emphasize that the condition is joint normality (or more generally, that the joint distribution is determined by its mean and covariance, as in elliptical distributions).

5. Connect to practical implications

Discuss why this matters in data science and finance, e.g., risk modeling, and caution against assuming independence without checking the joint distribution.

Key Points to Mention

  • Correlation measures linear dependence only.
  • Independence implies zero correlation, but not vice versa.
  • Counterexample: X ~ N(0,1), Y = X^2, then Cov(X,Y)=0 but Y is dependent on X.
  • Joint normality is a sufficient condition for zero correlation to imply independence.
  • In finance, returns are often assumed normal but may have fat tails, so zero correlation may not imply independence.
  • Elliptical distributions also have the property that zero correlation implies independence.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q5

Explain linear regression and walk through the assumptions behind OLS.

Technical Trade-offsData Modeling
Author's notes

Linearity, independence of errors, homoscedasticity, normality of errors, no perfect multicollinearity.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining linear regression in simple terms, then explain the OLS method and its objective. Systematically walk through the assumptions, emphasizing how violations can be detected and addressed, and relate them to practical implications in a banking context.

Pro tip: Mention that in finance, assumptions like no autocorrelation and homoscedasticity are often violated due to time series data, and briefly discuss robust alternatives like Newey-West standard errors to show depth.

1. Define Linear Regression

Explain that linear regression models the relationship between a dependent variable and one or more independent variables by fitting a linear equation.

2. Explain OLS

Describe OLS as a method that estimates coefficients by minimizing the sum of squared residuals, and mention its closed-form solution.

3. List Assumptions

Enumerate the key assumptions: linearity, independence, homoscedasticity, normality of errors, no multicollinearity, and no endogeneity.

4. Discuss Violations and Remedies

For each assumption, briefly explain how to detect violations (e.g., residual plots, VIF) and potential fixes (e.g., transformations, robust standard errors).

5. Relate to Banking Context

Highlight why these assumptions matter in banking, such as for risk modeling, and mention common challenges like non-normality in financial data.

Key Points to Mention

  • OLS minimizes the sum of squared residuals.
  • Assumptions: linearity, independence of errors, homoscedasticity, normality of errors, no multicollinearity, no endogeneity.
  • Violations can lead to biased or inefficient estimates.
  • Detection methods: residual plots, Durbin-Watson, VIF, Breusch-Pagan test.
  • Remedies: transformations, robust standard errors, regularization, instrumental variables.
  • In banking, assumptions like independence and homoscedasticity are often violated in time series data.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q6

What is the closed-form solution for the OLS estimator?

Data ModelingAlgorithms & Data Structures
Author's notes

Beta equals (X transpose X) inverse times X transpose Y.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by stating the closed-form solution clearly: β = (XᵀX)⁻¹Xᵀy. Then explain the intuition behind it—minimizing the sum of squared residuals—and discuss the assumptions and practical considerations, especially in a banking context where data may be large or collinear.

Pro tip: Mention that in practice, you rarely compute the inverse directly; instead, you use QR decomposition or gradient descent for numerical stability and scalability. This shows you understand both theory and real-world implementation.

1. State the formula

Clearly write the closed-form OLS solution: β = (XᵀX)⁻¹Xᵀy, defining X as the design matrix and y as the target vector.

2. Explain the derivation intuition

Describe that OLS minimizes the residual sum of squares (RSS), and setting the gradient to zero yields the normal equations, leading to the closed-form solution.

3. Discuss assumptions and conditions

Mention key assumptions: linearity, independence, homoscedasticity, normality of errors, and no perfect multicollinearity (XᵀX invertible).

4. Address practical considerations

Note that computing the inverse can be unstable or slow for large datasets; alternatives like QR decomposition, SVD, or regularized regression (ridge) are often used.

5. Relate to banking context

Connect to Bank of America by mentioning applications like credit scoring, risk modeling, or fraud detection, where interpretability and computational efficiency matter.

Key Points to Mention

  • The formula β = (XᵀX)⁻¹Xᵀy and its derivation from minimizing RSS.
  • Assumptions: linearity, independence, homoscedasticity, normality, no multicollinearity.
  • Numerical stability: why direct inversion is avoided; use QR or SVD.
  • Regularization: ridge regression as a solution when XᵀX is singular or ill-conditioned.
  • Computational complexity: O(n³) for inversion, making it impractical for large datasets.
  • Interpretability: OLS coefficients are easy to explain, which is valuable in regulated industries like banking.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q7

Why is multicollinearity a problem in regression, and how do you detect and address it?

Data ModelingTechnical Trade-offsRoot Cause Analysis
Author's notes

Multicollinearity inflates the variance of coefficient estimates, making them unstable and hard to interpret even if predictions are okay.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining multicollinearity and explaining its consequences on coefficient estimates and model interpretability. Then outline a systematic approach to detect it using VIF, correlation matrices, and condition indices, and finally discuss remedies such as removing variables, combining them, or using regularization. Emphasize the importance of context and business impact, especially in banking where interpretability and stability are critical.

Pro tip: In banking, multicollinearity can make it hard to explain which factors drive risk, so always tie your solution back to model interpretability and regulatory requirements. Mention that sometimes keeping correlated variables is acceptable if prediction is the only goal, but for inference, you must address it.

1. Define and explain the problem

Define multicollinearity as high correlation among predictors, leading to unstable coefficient estimates, inflated standard errors, and reduced interpretability. Explain why this matters for inference and business decisions.

2. Detect multicollinearity

Use correlation matrices to spot pairwise correlations, calculate Variance Inflation Factor (VIF) for each predictor (VIF > 5 or 10 indicates concern), and consider condition indices for broader diagnostics.

3. Assess impact and context

Evaluate whether multicollinearity affects your specific goal: if prediction is key, it may be less critical; if inference or coefficient interpretation is needed, it must be addressed. Consider business implications and regulatory needs.

4. Address multicollinearity

Apply remedies such as removing one of the correlated variables, combining them via PCA or factor analysis, using regularization (ridge, lasso), or collecting more data. Choose based on model purpose and domain knowledge.

5. Validate and communicate

Re-check VIF and model stability after changes, validate with cross-validation, and clearly communicate the rationale and impact to stakeholders, ensuring transparency.

Key Points to Mention

  • Definition: high correlation among independent variables
  • Consequences: unstable coefficients, inflated standard errors, misleading p-values
  • Detection: correlation matrix, VIF, tolerance, condition index
  • Remedies: variable removal, PCA, ridge regression, lasso, domain-driven combination
  • Trade-offs: prediction vs. inference, bias-variance trade-off
  • Banking context: regulatory interpretability, risk model stability

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q8

Explain Ridge and Lasso regression and how they relate to the multicollinearity problem.

Technical Trade-offsData Modeling
Author's notes

Ridge adds an L2 penalty and shrinks coefficients toward zero without zeroing them out, which helps with collinearity by spreading the weight.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining Ridge and Lasso as regularized linear regression techniques, then explain how their penalties (L2 and L1) shrink coefficients differently. Connect this to multicollinearity by discussing how Ridge stabilizes estimates when predictors are correlated, while Lasso performs feature selection by zeroing out coefficients, which can arbitrarily drop correlated features. Emphasize the trade-offs and when to use each in a banking context.

Pro tip: Mention that in banking, interpretability and regulatory compliance often favor Lasso for its feature selection, but if all correlated features are important, Ridge is preferred. Also, note that Elastic Net combines both and can be a safe default.

1. Define Ridge and Lasso

Explain that both are linear regression variants with regularization: Ridge adds an L2 penalty (sum of squared coefficients), Lasso adds an L1 penalty (sum of absolute coefficients).

2. Explain the effect on coefficients

Describe how Ridge shrinks coefficients towards zero but never exactly zero, while Lasso can shrink some coefficients exactly to zero, performing feature selection.

3. Define multicollinearity and its problems

Define multicollinearity as high correlation among predictors, which inflates coefficient variance and makes estimates unstable and hard to interpret.

4. Relate Ridge and Lasso to multicollinearity

Explain that Ridge handles multicollinearity by distributing coefficient weights among correlated features, reducing variance. Lasso, however, tends to pick one correlated feature and discard others, which can be problematic if all are relevant.

5. Discuss trade-offs and practical considerations

Highlight that Ridge is better when all features contribute, Lasso when sparse solutions are desired. Mention Elastic Net as a compromise and the importance of cross-validation for tuning lambda.

Key Points to Mention

  • L2 vs L1 penalty and their geometric interpretations
  • Bias-variance trade-off: regularization increases bias but reduces variance
  • Multicollinearity causes unstable coefficients and inflated standard errors
  • Ridge's grouping effect: correlated features get similar coefficients
  • Lasso's feature selection: can arbitrarily choose one among correlated features
  • Elastic Net as a combination of L1 and L2 penalties
  • Use cross-validation to select the regularization parameter lambda
  • Banking context: interpretability and regulatory requirements may influence choice

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q9

Explain PCA, including what eigenvalues and eigenvectors represent in that context, and what the main limitations are.

Data ModelingTechnical Trade-offs
Author's notes

PCA finds directions of maximum variance in the data.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start with a clear, intuitive definition of PCA as a dimensionality reduction technique that finds new orthogonal axes (principal components) maximizing variance. Then explain eigenvalues and eigenvectors as the variance magnitude and direction of these components, and finish by discussing limitations such as interpretability, linearity, and sensitivity to scaling.

Pro tip: Relate PCA to a real-world banking scenario, like reducing the dimensionality of customer transaction data for credit risk modeling, to show practical understanding and business relevance.

1. Define PCA

Explain PCA as a method to reduce dimensionality by projecting data onto a lower-dimensional subspace that captures the most variance.

2. Explain Eigenvectors and Eigenvalues

Describe eigenvectors as the directions of the new axes (principal components) and eigenvalues as the amount of variance explained by each component.

3. Describe the PCA Process

Outline the steps: standardize data, compute covariance matrix, find eigenvectors/eigenvalues, sort by eigenvalues, and project data onto top components.

4. Discuss Limitations

Mention key limitations: assumes linearity, may lose interpretability, sensitive to scaling, and can discard useful information if variance is not aligned with predictive power.

5. Conclude with Practical Implications

Summarize when PCA is useful (e.g., high-dimensional data, noise reduction) and when to consider alternatives (e.g., nonlinear methods like t-SNE or autoencoders).

Key Points to Mention

  • PCA is an unsupervised linear dimensionality reduction technique.
  • Eigenvectors represent the directions of maximum variance (principal components).
  • Eigenvalues quantify the variance captured by each principal component.
  • PCA assumes linear relationships and may not capture complex nonlinear patterns.
  • Standardization is crucial because PCA is sensitive to feature scales.
  • Interpretability of principal components can be challenging in business contexts.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.