Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 8 views

    Worksheet 2

    Uploaded by

    icelon N
    This document provides instructions for a computer experiments and uncertainty quantification worksheet involving a regression model. Students are asked to: 1. Generate data from the given model and plot the results. 2. Fit a polynomial regression to the data and analyze the model fit. 3. Assess the prediction error of the polynomial regression model using different degrees of polynomials on a test dataset.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Worksheet 2

    Uploaded by

    icelon N
    8 views2 pages
    This document provides instructions for a computer experiments and uncertainty quantification worksheet involving a regression model. Students are asked to: 1. Generate data from the given model and plot the results. 2. Fit a polynomial regression to the data and analyze the model fit. 3. Assess the prediction error of the polynomial regression model using different degrees of polynomials on a test dataset.

    Original Title

    Worksheet2

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides instructions for a computer experiments and uncertainty quantification worksheet involving a regression model. Students are asked to: 1. Generate data from the given model and plot the results. 2. Fit a polynomial regression to the data and analyze the model fit. 3. Assess the prediction error of the polynomial regression model using different degrees of polynomials on a test dataset.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    8 views2 pages

    Worksheet 2

    Uploaded by

    icelon N
    This document provides instructions for a computer experiments and uncertainty quantification worksheet involving a regression model. Students are asked to: 1. Generate data from the given model and plot the results. 2. Fit a polynomial regression to the data and analyze the model fit. 3. Assess the prediction error of the polynomial regression model using different degrees of polynomials on a test dataset.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Winter 2023 NCKU Modular Course

    Computer Experiments and Uncertainty Quantification


    Worksheet 2
    Due: 5 pm of Jan 11 2023
    Consider the following model, which is denoted by model (1):
    𝑦 = 𝑓(𝑥) + 𝑒rror,
    𝑓(𝑥) = 10 + 10 sin(4𝜋𝑥 0.9) 𝑒 −1.5𝑥 , (1)
    𝑒rror ~ 𝑁(0, 1)

    I. Generate a dataset
    a. Generate 11 equally spaced points between 0 and 1 and assign
    the resulting vector as a variable x_n11. Please type set.seed(2)
    before you start writing your results.

    b. Write a function for 𝑓(𝑥 ) and name the function by CE_function1.

    c. Generate 11 random error values from normal distribution with


    mean 0 and variance 1 and save it as a vector with name
    error_n11.

    d. Use results from a and c to generate 11 values as a vector for y


    and assign the results as a variable y_n11.

    e. Create a dataframe which includes two columns x_n11 and y_n11


    and assign the resulting dataframe as a variable data_n11 .

    II. Visualize the data


    f. b. Plot y_n11 versus x_n11. Set the title as “Scatter Plot 1”. (Hint:
    Check function plot in R )

    III. Fit the model


    g. Fit a polynomial regression model to the dataset data_n11 by
    using the following code

    #sample R code for question f


    lm_n11 <- lm(y_n11 ~ poly(x_n11, 3), data_n11)
    Hint: (1) Please check function lm and poly in R and (2) The example
    fits a polynomial regression with degrees 3.

    h. Try summary(lm_n11) and describe what you observed.

    i. Plot the residuals (observed value – fitted value) versus x_n11?


    What do you observe? Hint: the fitted values can be generated by
    predict(lm_n11, data_n11)

    j. Plot the residuals versus the fitted values.

    k. Report the R square values and adjusted R square values of


    lm_n11 in question (g) by using the following code:
    summary(lm_n11)$r.squared
    summary(lm_n11)$adj.r.squared

    Check google and then provide your interpretation the r.squared


    value and the adj.r.squared value you reported.

    IV. Assess Prediction Error


    l. Generate a testing dataset with 20 data points from model (1) and
    write a for loop in R that can assess the prediction errors of the
    polynomial regression model in (g) with degrees 1, 2, 3, 4, 5, 6.
    Save the prediction errors in a list format.

    Guideline for the worksheet


    1. Please use R markdown to generate one pdf file as your worksheet
    results and submit it to Moodle, so you can include your R code.
    Note: It is not necessary to type the question in your results for the
    submission.
    2. Please remember to type the question number with your answer and
    R code.

    You might also like