Preview Questions Data Analyst | Role Specific Skill

Question 1: How do you design a report that accurately represents financial trends over time while accounting for seasonality?

Which action should you take?

Choose only one option

I report the data without considering time-based fluctuations

I ignore seasonality and focus on raw data trends

I use simple averages without adjustments for seasonality

I use decomposition techniques or time series models to adjust for seasonality, ensuring trends are displayed clearly

Question 2: In a classification model, what is the purpose of using ROC curves and AUC scores for BFSI data analysis?

Which action should you take?

Choose only one option

To check for multicollinearity

To evaluate the strength of linear relationships

To measure the prediction performance of a classifier

To assess the normality of data distribution

Question 3: What is the primary concern when handling timestamp data in BFSI datasets during preprocessing?

Which action should you take?

Choose only one option

Removing duplicate timestamps

Removing any time-related outliers

Ensuring the data is in a consistent format

Imputing missing timestamp data

Question 4: What should you do when you discover that a financial dataset contains incorrect data types (e.g., numbers stored as text)?

Which action should you take?

Choose only one option

Ignore the issue and proceed with the analysis

Remove the entire column from the dataset

Convert the columns to the correct data type using a function

Manually correct each record

Question 5: How do you identify long-term trends in a financial dataset with fluctuating seasonal behavior, like loan approval rates?

Which action should you take?

Choose only one option

I apply a linear regression model and assume all changes are trend-based

I use simple moving averages and ignore seasonality

I focus only on monthly data without considering long-term variations

I apply a time series decomposition to separate trend, seasonality, and noise, and then analyze the trend component

Question 6: How do you evaluate the effectiveness of a risk model built to predict loan defaults in a highly imbalanced dataset?

Which action should you take?

Choose only one option

I would use metrics such as ROC-AUC, precision-recall curves, and cross-validation to assess model performance

I would rely on F1 score and ignore other metrics

I would evaluate based solely on the number of correct predictions

I would only evaluate the model using accuracy, as it is the simplest metric