Will someone answer this question for me please. All I need are the answers noth

Question

Will someone answer this question for me please. All I need are the answers nothing extra. Thankyou

4. Solution Concentration: This is a new data set and is in columns C1 and C2 in hw4da?a,my. A chemist studied the concentration of a solution ?55ng) over time (X-time). 15 identical solutions were prepared and randomly divided into 5 sets of 3. These 5 sets were measured after 1,3,5,7, and 9 hours and provide in temporal order. a) Examine a scatter plot of Y-conc vs. X-time. Does the relationship appear linear or not? Construct a linear regression model in Minitab using X-time to predict Y-conc and obtain the residual vs. fit plot. Based on this plot, do you believe this is a valid linear model? Obtain a normal probability of residuals. Based on this plot, and this plot, do you believe this is valid linear model? b) c) d) Based on your overall examination of the diagnostics, determine which variable should be considered for transformation. Note: Because time is our predictor variable, we might expect some autocorrelation among the residuals. As it turns out, this model fails the D-W test and autocorrelation is indeed present. We wont learn how to correct this problem until STA-590. For now, we will ignore it. You should, however, be aware that autocorrelation may have an effect on the normality of the residuals. Sometimes knowing what may be causing a problem is as important as detecting the problem existing in the first place

Explanation / Answer

a) Scatter Plot :

From the scatter plot we see that all data points are not lie in one line ( either increasing or decreasing).

So we say that there is no linear relationship .

-------------------------------------------------------------------------------

b)

Linear regression model : Y -conc = 2.575 - 0.3240 XTime

Residual Vs fitted Plot :

If you find equally spread residuals around a horizontal line without distinct patterns, that is a good indication you don’t have non-linear relationships.

So here from Residual vs fitted value plot we can't see any distinct pattern.

So we say that there is no linear relation .

----------------------------------------------------------------------

c )

Normalm probability plout for residual :

It’s good if residuals are lined well on the straight dashed line.

So here some residual not strickly lie to the line . Since this is not linear model .

Related Questions

Navigate

• Browse (All)
• Browse W
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.