Preview Questions Data Analyst | Role Specific Skill

Question 1: In the context of BFSI, what is the primary purpose of data warehousing, and how does it differ from operational databases?

Which action should you take?

Choose only one option

To handle real-time data processing

To aggregate and store historical data for analysis and reporting

To replace operational databases entirely

To store transactional data for daily operations

Question 2: In the context of BFSI, which data normalization technique is most effective for handling skewed financial data distributions before applying machine learning algorithms?

Which action should you take?

Choose only one option

Min-Max Scaling

Robust Scaling

Z-Score Standardization

Log Transformation

Question 3: Which statistical distribution is most commonly used to model the frequency of rare events, such as loan defaults?

Which action should you take?

Choose only one option

Poisson distribution

Log-normal distribution

Binomial distribution

Normal distribution

Question 4: How do you ensure that a financial model is not overfitting to historical data when making predictions for future market behavior?

Which action should you take?

Choose only one option

I would use cross-validation, regularization techniques (Lasso/Ridge), and assess model performance on out-of-sample data

I would use more training data but ignore validation sets

I would increase the number of predictors in the model to improve accuracy

I would select the model based only on its performance on training data

Question 5: Which statistical test is appropriate for comparing the means of three or more independent groups (e.g., financial products in different markets)?

Which action should you take?

Choose only one option

t-test

Chi-square test

ANOVA

Pearson's correlation

Question 6: When performing time series analysis, what is the main purpose of decomposition?

Which action should you take?

Choose only one option

To reduce noise and improve model accuracy

To identify trends, seasonality, and residual components

To smooth out the data for easier analysis

To predict future values using autoregression