Master in Data Science (SMS, TU) Statistical Computing with R Question Paper 2080 Nepal

Describe supervised learning with focus on:

(a) Grammar of graphics (b) Layers in grammar of graphics (c) Statistical transformations in grammar of graphics

Describe supervised learning linear regression model with focus on:

(a) Pre-requisites before fitting this type of model (b) Multicollinearity and its importance in the multiple linear regression model (c) Best regularized regression model to control the multicollinearity in multiple linear regression model

Describe supervised learning classification regression model with focus on:

(a) Model fit indices (b) Confusion matrix (c) Prediction accuracy indices

Describe supervised learning method with focus on:

(a) Single layer, feed-forward neural network (b) Activation functions used in the neural network models (c) Network model with input, hidden and output layers

Describe decision tree classification model with focus on:

(a) Bagging (b) Improved bagging (c) Boosting

Do the following in R Studio using ggplot2 package with R script:

(a) Create a dataset with following variables: age (18-99 years), sex (male/female), educational levels (No education/Primary/Secondary/Beyond secondary), socio-economic status (Low, Middle, High) and body mass index (14 – 38) with random 100 cases of each variable. Your roll number must be used to set the random seed. (b) Create a line chart of age variable using ggplot2 package and interpret the result carefully (c) Create scatter plot of age and body mass index variables using ggplot2 package and interpret the result carefully. (d) Create classes of body mass index variable as: <18, 18-24, 25-30, 30+ and show it as pie chart using ggplot2 package and interpret it carefully (e) Create classes of age variable as <15, 15-59 and 60+ and show it as bar diagram using the ggplot2 package and interpret it carefully

Do the followings in R Studio using "Bfox" dataset with R script:

(a) Divide the Bfox data into train and test datasets with 70:30 random splits (b) Fit a supervised linear regression model and KNN regression model on train data with "debt" as dependent variable and all other variables as independent variable (c) Get the summary of the model, fit indices and interpret them carefully (d) Predict the debt variable in the test data and interpret the result carefully (e) Get the fit indices (R-square, MSE, RMSE) of the predicted model and interpret them carefully

Do the following in R Studio using "Arrests" dataset with R script:

(a) Divide the mtcars data into train and test datasets with 80:20 random splits (b) Fit a supervised logistic regression model and naïve bayes classification models on train data with "released" as dependent variable and colour, age, sex, employed and citizen as independent variable (c) Get the confusion matrix, sensitivity, specificity of the fitted model and interpret them carefully (d) Predict the transmission variable in the test data and interpret the result carefully (e) Get the confusion matrix, sensitivity, specificity of the predicted model and interpret them carefully

Do as follows using "mtcars" dataset with R script:

(a) Create a "car scale" with all the variables variables using the Principal Component Analysis (PCA) model (b) Interpret the results of the PCA carefully in terms of Kaiser's criteria (c) Get scree plot and decide the number of components to be retained (d) Get the bi-plot of the fitted model and interpret it carefully (e) Improve the fitted model with VARIMAX process and interpret the results carefully

OR

Do as follows using "USArrests" dataset in R studio with R script:

(a) Get dissimilarity distance as state. dissimilarity object (b) Fit a classical multidimensional model using the state. Dissimilarity object (c) Get the summary of the model and interpret it carefully (d) Get the plot of the model and interpret it carefully (e) Compare this model with the first two components from principal component analysis model in this data

Use the "mtcars" data and do as follows with R Script:

(a) Fit a hierarchical clustering model using single linkage and get the dendogram for this model (b) Fit a hierarchical clustering model using complete linkage and get the dendogram for this model (c) Fit a hierarchical clustering model using average linkage and get the dendogram for this model (d) Show the number of clusters (k) to retain for the data using ablines in the dendogram of the best model (e) Get the best value of number of clusters to form (k) using the fitted model above

OR

Load the "USArrests" data file into R Studio and do as follows with R script:

(a) Fit a k-means clustering model in the data with k=2 and 3 (b) Get the cluster means and within sum of square value each model and interpret them carefully (c) Plot the clusters formed by k=2 and 3 and interpret them carefully (d) Add cluster centers for plot with k=3 and interpret it carefully (e) Visualize the clusters for k=3 and interpret it carefully