91Ó°ÊÓ

The presence of hard alloy carbides in high chromium white iron alloys results in excellent abrasion resistance, making them suitable for materials handling in the mining and materials processing industries. The accompanying data on \(x=\) retained austenite content \((\%)\) and \(y=\) abrasive wear loss \(\left(\mathrm{mm}^{3}\right)\) in pin wear tests with garnet as the abrasive was read from a plot in the article "Microstructure-Property Relationships in High Chromium White Iron Alloys" (Internat. Mater. Rev., 1996: 59–82). $$ \begin{aligned} &\begin{array}{c|ccccccccc} x & 4.6 & 17.0 & 17.4 & 18.0 & 18.5 & 22.4 & 26.5 & 30.0 & 34.0 \\ \hline y & .66 & .92 & 1.45 & 1.03 & .70 & .73 & 1.20 & .80 & .91 \end{array}\\\ &\begin{array}{l|llllllll} x & 38.8 & 48.2 & 63.5 & 65.8 & 73.9 & 77.2 & 79.8 & 84.0 \\ \hline y & 1.19 & 1.15 & 1.12 & 1.37 & 1.45 & 1.50 & 1.36 & 1.29 \end{array} \end{aligned} $$ a. What proportion of observed variation in wear loss can be attributed to the simple linear regression model relationship? b. What is the value of the sample correlation coefficient? c. Test the utility of the simple linear regression model using \(\alpha=.01\). d. Estimate the true average wear loss when content is \(50 \%\) and do so in a way that conveys information about reliability and precision. e. What value of wear loss would you predict when content is \(30 \%\), and what is the value of the corresponding residual?

Step by step solution

01

Organize Data

First, we arrange the given data points, representing retained austenite percentages \(x\) and abrasive wear loss \(y\). Using this data, we aim to establish a simple linear regression model that predicts \(y\) based on \(x\).

02

Calculate Regression Model

Perform a simple linear regression analysis to model the relationship between \(x\) (retained austenite content) and \(y\) (abrasive wear loss). The regression line will have the form \(y = b_0 + b_1x\), where \(b_0\) and \(b_1\) are the intercept and slope, respectively.

03

Compute Coefficient of Determination

Calculate the coefficient of determination \(R^2\), which tells us the proportion of variation in \(y\) that is explained by the variation in \(x\). Use the formula \(R^2 = 1 - \frac{SS_{res}}{SS_{tot}}\), where \(SS_{res}\) is the residual sum of squares and \(SS_{tot}\) is the total sum of squares.

04

Find Sample Correlation Coefficient

Using the formula \(R = \sqrt{R^2}\) (with the same sign as \(b_1\)) to find the sample correlation coefficient, which measures the strength and direction of the linear relationship between \(x\) and \(y\).

05

Perform Hypothesis Test

Conduct a hypothesis test for the slope \(b_1\) using \(t\)-test to assess the utility of the regression model. The null hypothesis \(H_0\) states that \(b_1 = 0\). Compare the \(t\)-value with the critical \(t\)-value at \(\alpha = 0.01\) to make a decision.

06

Estimate and Predict Wear Loss

Use the regression equation to estimate the wear loss when \(x = 50\%\) and predict wear loss at \(x = 30\%\). For \(x = 50\%\), compute a confidence interval to convey precision. For \(x = 30\%\), calculate the residual by subtracting the predicted \(y\) from the observed \(y\).

Unlock Step-by-Step Solutions & Ace Your Exams!

• Full Textbook Solutions

  Get detailed explanations and key concepts

• Unlimited Al creation

  Al flashcards, explanations, exams and more...

• Ads-free access

  To over 500 millions flashcards

• Money-back guarantee

  We refund you if you fail your exam.

Over 30 million students worldwide already upgrade their learning with 91Ó°ÊÓ!

Key Concepts

These are the key concepts you need to understand to accurately answer the question.

Coefficient of Determination

The Coefficient of Determination, denoted as \( R^2 \), is a key statistic in evaluating the performance of a simple linear regression model. It provides a measure of how well the regression line approximates real data points.
The value of \( R^2 \) ranges from 0 to 1. An \( R^2 \) value closer to 1 indicates a strong and perfect fit, where all data points lie on the regression line, reflecting that the model explains a great portion of the variability in the dependent variable \( y \). Conversely, a smaller \( R^2 \) value suggests that the model does not explain much of the variability in \( y \).
In practice, you calculate \( R^2 \) using the formula \( R^2 = 1 - \frac{SS_{res}}{SS_{tot}} \), where \( SS_{res} \) is the sum of squares of residuals and \( SS_{tot} \) is the total sum of squares. From this, students can see the amount of variability captured by the model in relation to the total variability present in the system.

Sample Correlation Coefficient

The Sample Correlation Coefficient, represented by \( R \), measures the strength and direction of the linear relationship between two variables in a dataset, namely \( x \) and \( y \). It's derived from the Coefficient of Determination, \( R^2 \), by taking the square root \( R = \sqrt{R^2} \).
The correlation coefficient ranges from -1 to 1.

• A value of 1 indicates a perfect positive linear relationship.
• A value of -1 signifies a perfect negative linear relationship.
• A value of 0 implies no linear relationship between the variables.

In our abrasion wear model, for example, a high positive \( R \) value would suggest that increases in the percentage of retained austenite content are generally associated with increases in wear loss, whereas a negative value would indicate the opposite.

Hypothesis Testing

Hypothesis Testing in simple linear regression revolves around determining the statistical significance of the slope coefficient, \( b_1 \), which represents the rate of change of the dependent variable \( y \) with respect to \( x \).
The null hypothesis \( H_0: b_1 = 0 \) assumes that there is no relationship between \( x \) and \( y \). To check the validity of this hypothesis, a \( t \)-test is performed.

Here's how you do it:

• Calculate the \( t \)-statistic using sample data.
• Compare this \( t \)-value to the critical value from a \( t \)-distribution table at a significance level \( \alpha \), such as 0.01 (indicating a 1% risk of concluding that a relationship exists when there is none).
• If \( |t| \) exceeds the critical value, you reject the null hypothesis, suggesting that the regression slope is significantly different from zero and the model is useful.

This process helps determine whether the change in \( x \) truly affects the level of abrasive wear loss observed.

Abrasive Wear Loss Prediction

Abrasive Wear Loss Prediction involves using the established regression model equation to forecast wear loss for specific austenite content values. This is particularly useful in the materials industry to preemptively assess the wear resistance of materials.
To predict wear loss when \( x = 30\% \) for instance, you substitute this value into your regression equation to find the expected \( y \).
Moreover, predictions are made more reliable when supplemented with confidence intervals, which provide a range within which the true wear loss is likely to fall. For example, predicting wear loss when the content is \( 50\% \) would include calculating a confidence interval to understand the precision of this prediction.
Finally, when considering residuals - the difference between observed and predicted \( y \) values - it helps evaluate how well your model fits specific data points. A small residual suggests a good fit, while a large one indicates the model might need refinement or reevaluation. This is crucial for ensuring the accuracy and effectiveness of predictions in real-world scenarios.