MCQs on Statistical Modeling | R
Statistical modeling in R is a critical skill for data analysis, allowing for predictive insights. This chapter covers key topics like linear and logistic regression, time series analysis, ARIMA models, and model validation.
Chapter: Statistical Modeling in R – MCQs
1. Linear and Logistic Regression
- What is the purpose of linear regression in R?
- a) To predict categorical outcomes
- b) To predict continuous outcomes
- c) To calculate the correlation between variables
- d) To cluster data
- Which function in R is used for fitting a linear regression model?
- a)
lm() - b)
logit() - c)
glm() - d)
fit()
- a)
- What is the assumption underlying linear regression?
- a) The relationship between variables is nonlinear
- b) The residuals are normally distributed
- c) The data must be categorical
- d) The predictors must be independent
- Which of the following is true about logistic regression?
- a) It is used for predicting continuous variables
- b) It is used for predicting categorical variables
- c) It assumes a normal distribution of errors
- d) It does not require any transformations of the dependent variable
- What is the main output of a logistic regression model in R?
- a) Coefficients and odds ratios
- b) Predicted probabilities
- c) Residuals
- d) AUC score
- Which of the following is NOT a valid link function in logistic regression?
- a) Logit
- b) Probit
- c) Identity
- d) Exponential
- What is the interpretation of the coefficients in logistic regression?
- a) They represent the mean difference between groups
- b) They are the probabilities of the outcome event
- c) They represent the odds ratio for a one-unit change in the predictor
- d) They represent the R-squared value
- What does the “Null Deviance” in a logistic regression output represent?
- a) The deviance of a model with only the intercept
- b) The deviance of the fitted model
- c) The predicted value of the model
- d) The variance of the residuals
- How do you assess the goodness of fit in logistic regression?
- a) Using R-squared
- b) Using deviance or AIC
- c) Using correlation coefficient
- d) Using MSE (Mean Squared Error)
- What is the purpose of multicollinearity in linear regression?
- a) It improves the model
- b) It causes biased estimates of coefficients
- c) It helps in model selection
- d) It is not a concern in linear regression
2. Time Series Analysis and ARIMA Models
- What is a time series model used for?
- a) To predict future values based on historical data
- b) To cluster time-based data
- c) To reduce dimensionality of time data
- d) To visualize trends in time series
- What does ARIMA stand for?
- a) Autoregressive Integrated Moving Average
- b) Autoregressive Interpolation Moving Average
- c) Automated Regression Integrated Moving Average
- d) Autoregressive Independent Moving Average
- Which component of ARIMA is responsible for the autoregressive part?
- a)
AR - b)
MA - c)
I - d)
T
- a)
- In an ARIMA model, what does the “I” stand for?
- a) Interaction
- b) Integration
- c) Inference
- d) Indicator
- What is the primary purpose of differencing in ARIMA models?
- a) To transform the data into a stationary series
- b) To reduce noise in the data
- c) To adjust for seasonal effects
- d) To estimate the model’s accuracy
- Which of the following is NOT a component of a typical time series?
- a) Trend
- b) Seasonal
- c) Irregular
- d) Logistic
- Which function is used to fit an ARIMA model in R?
- a)
auto.arima() - b)
arima() - c)
tslm() - d)
lm()
- a)
- How do you check the stationarity of a time series in R?
- a) Using the Augmented Dickey-Fuller (ADF) test
- b) Using the Shapiro-Wilk test
- c) By visualizing the series only
- d) Using the chi-squared test
- What is a common diagnostic plot used to assess the residuals in an ARIMA model?
- a) Histogram
- b) Q-Q plot
- c) Autocorrelation Function (ACF) plot
- d) Boxplot
- What does the ACF (Autocorrelation Function) plot show?
- a) The relationship between the residuals and fitted values
- b) The correlation between observations at different time lags
- c) The variance of the time series
- d) The distribution of the residuals
3. Model Validation and Diagnostics
- Which of the following is used to assess the accuracy of a regression model in R?
- a) R-squared
- b) p-value
- c) Coefficients
- d) p-value
- What is cross-validation used for in statistical modeling?
- a) To evaluate model performance using a subset of data
- b) To estimate coefficients
- c) To adjust for outliers
- d) To check for multicollinearity
- Which metric is commonly used for model evaluation in classification tasks?
- a) Mean Squared Error (MSE)
- b) Confusion matrix
- c) R-squared
- d) F-test
- What is the purpose of the residuals in a regression model?
- a) To estimate the error term
- b) To check the assumption of homoscedasticity
- c) To evaluate model accuracy
- d) To check for overfitting
- What is the assumption of homoscedasticity in regression analysis?
- a) The variance of the errors is constant across all levels of the independent variable
- b) The errors are normally distributed
- c) The data must be uncorrelated
- d) The errors should follow a linear trend
- Which function is used to calculate the Variance Inflation Factor (VIF) in R?
- a)
vif() - b)
vif.lm() - c)
vif.stats() - d)
calc_vif()
- a)
- Which diagnostic plot is used to check for normality in residuals?
- a) Residual plot
- b) Q-Q plot
- c) ACF plot
- d) Histogram plot
- What is the purpose of the p-value in statistical models?
- a) To indicate the significance of the model coefficients
- b) To estimate the variance
- c) To check for linearity
- d) To assess the quality of residuals
- Which test is used to detect multicollinearity in regression models?
- a) VIF (Variance Inflation Factor)
- b) Shapiro-Wilk test
- c) t-test
- d) Chi-squared test
- How do you assess the overfitting of a model?
- a) By using the cross-validation technique
- b) By checking residuals only
- c) By adjusting the model parameters manually
- d) By visualizing the data
Answers Table
| Qno | Answer |
|---|---|
| 1 | b) To predict continuous outcomes |
| 2 | a) lm() |
| 3 | b) The residuals are normally distributed |
| 4 | b) It is used for predicting categorical variables |
| 5 | a) Coefficients and odds ratios |
| 6 | c) Identity |
| 7 | c) They represent the odds ratio for a one-unit change in the predictor |
| 8 | a) The deviance of a model with only the intercept |
| 9 | b) Using deviance or AIC |
| 10 | b) It causes biased estimates of coefficients |
| 11 | a) To predict future values based on historical data |
| 12 | a) Autoregressive Integrated Moving Average |
| 13 | a) AR |
| 14 | b) Integration |
| 15 | a) To transform the data into a stationary series |
| 16 | d) Logistic |
| 17 | a) auto.arima() |
| 18 | a) Using the Augmented Dickey-Fuller (ADF) test |
| 19 | c) Autocorrelation Function (ACF) plot |
| 20 | b) The correlation between observations at different time lags |
| 21 | a) R-squared |
| 22 | a) To evaluate model performance using a subset of data |
| 23 | b) Confusion matrix |
| 24 | a) To estimate the error term |
| 25 | a) The variance of the errors is constant across all levels of the independent variable |
| 26 | a) vif() |
| 27 | b) Q-Q plot |
| 28 | a) To indicate the significance of the model coefficients |
| 29 | a) VIF (Variance Inflation Factor) |
| 30 | a) By using the cross-validation technique |