代写R做prediction 编程、代写ML method

日期：2018-04-24 07:31

Statistics Project – Resistors

For this project we will assume that we tested two batches of resistors (in ohms) from two different

plants, A and B. The first batch was very small. The second batch was much larger. This project is going

to be done entirely in MatLab. Select a member to email your files following the instructions below by

11:59PM, Friday, April 27

th

. This member MUST copy all other members in the email so they know that

the file was sent and when. All files that you need are to be found in the folder named “Project” on the

main Moodle page.

First, you must retrieve your data. This can be found in the file, “STAT383_project_data.xlsx”. In this file,

you will see two columns of data: Data A and Data B. Press F9 to randomize the data and copy only the

data points of the column marked “Data A” and paste them into a txt file (I would use Notepad or

something) and name it “dataA.txt”. Next, do the same thing for Data B in a separate txt file. I have

already set up the code to import the data files for you.

NOTE: If you see something at the top of Excel like the following,

you can simply select “Enable Editing” so that you can get the correct data sets. The data sets will be

unique for each group.

Next, open the file that is called, “STAT383_Project_lastname_studentID.m” . Replace “lastname” and

“student ID” with your correct values. This is where you are going to put your code. At the very top,

there is a place for you to type in your FULL NAME and your STUDENT ID. Make sure it stays commented

out.

% Student Names: Your FULL NAME(s)

% Student IDs: Your STUDENT ID(s)

Next there will be a section called %% Init. This section contains some code to clear up your

workspace, etc., before you run your code. Also, at the bottom is something that looks like this,

disp(['Code by: FULL NAME STUDENT ID'])

disp(['Code by: FULL NAME STUDENT ID'])

disp(['Code by: FULL NAME STUDENT ID'])

disp([' ']);

The command disp will display the words between ([‘ … ‘]) in the command window. Here, you will

again put your FULL NAME and STUDENT ID and also the FULL NAME of those you worked with. You will

work in a group of three people. (You can use “fprintf” if you like instead.)

Following this code, you will see there is code to import the data is into two arrays: dataA and dataB. It

is important that you import the data as directed above and in the code so that I can run your data with

my code to check for the correct answers.

Next, you are asked to make four plots: A box or whisker plot, and a histogram for each data set. (Make

sure they are labeled appropriately.)

%% Plot the data (dataA)

% (Box or Whisker) For the key parts of the plot, see figure 6.24

% (Histogram)

Use something like ,

figure;boxplot(dataA);title ('Boxplot for DataA');ylabel('ohms')

So that each plot has a separate figure generated. After each set of plots (A and B), you will be asked the

following questions,

%% Questions

% Is the data skewed? Why or why not. Use your plots to support your

% analysis. HINT: Read 6.3 carefully.

You should answer the question in complete sentences and display it in the command window using the

disp command as above. Also, to display numerical solutions you can use the num2str command like this

(notice I also put a disp([' ']); to separate the sentences),

Y = 5

disp(['The value of Y is ' num2str(Y) ' and it is good.']);

disp([' ']);

Under the next section, %% Sample Analysis you will calculate the sample mean, mode, median

and sample standard deviation for each data set. The median you can get from the box plot or the

MatLab command median(data) and the mode may be seen from the data or use the MatLab command

mode(data). However, the sample mean and sample standard deviation MUST be calculated using the

formulas from class, i.e. do not use software commands like mean(X) but you can still use sum.

The rest of the instructions are in the m.file.

Read the commented text and follow the directions carefully. Ask if you any questions but ask sooner

than later.

For your project to be successful (i.e. full points), you need to send me ZIP file of your data files and your

MatLab file (m.file) . In addition, the following must be followed:

1) I should be able to run your file with no errors. This is easy to check before you send it and may

result in the loss of quite a few points if I cannot resolve your errors.

2) You must send me a ZIP file that contains your data files and your m.file code. Failure to do this

will result in ZERO credit.

3) Label your completed zip file – “STAT383_Project_lastname_studentID.zip” where you need to

include your last name and student ID. Failure to do this will result in ZERO credit.

Below is what I expect to see displayed in the command window when I run your code. The values will of

course be different. Also, below are the graphs properly labeled that your code should generate.

Code by: Adom Giffin 0123456

Code by: Adom Giffin 0123456

Code by: Adom Giffin 0123456

The data in the first data set, A, seem to be positively or right skewed. This is illustrated by the fact

that the length of the positive whisker is much longer, 2.8333 times more than the negative whisker.

Also, the histogram seems to have a positive skew.

The data in the second data set, B, seem not to be skewed. The whiskers seem to be roughly the

same length. The positive whisker is only 1 times more than the negative whisker. This is also

illustrated by the fact that the histogram seems to be fairly symmetric.

For data set A:

The sample mean is 98.85.

The mode is 101.

The median is 99.

The sample standard deviation is 4.4988.

For data set B:

The sample mean is 99.62.

The mode is 99.

The median is 100.

The sample standard deviation is 4.7605.

The two-tailed, 95% confidence interval for the sample mean of data set A, is (96.7445, 100.9555)

The two-tailed, 95% confidence interval for the sample mean of data set B, is (98.6754, 100.5646)

The two-tailed, 99% confidence interval for the sample mean of data set A, is (95.972, 101.728)

The two-tailed, 99% confidence interval for the sample mean of data set B, is (98.3697, 100.8703)

The T cdf for t = 2.262 is P(T<t) = 0.97499

For H0: mu_A = 98.25, the p-value = 0.55792

For H0: mu_B = 98.25, the p-value = 0.0049053

The p-value for dataA (0.55792) is greater than 0.10

therefore it is plausible.

The p-value for dataB (0.0049053) is less than 0.01

therefore it is not plausible.

The absolute value of the t statistic for dataA (0.59644) is less than the critical value, 2.093

and therefore we ACCEPT the null.

The absolute value of the t statistic for dataB (2.8779) is greater than the critical value, 1.9842

and therefore we REJECT the null.

For H0: muA - muC = 0, the paired p-value = 0.33459

The absolute value of the paired t statistic (0.99005) is less than the critical value, 2.093

and therefore we ACCEPT the null.

For H0: muA - muD = 0, the un-paired p-value = 0.47363

The absolute value of the un-paired t statistic (0.7244) is less than the critical value, 2.0301

and therefore we ACCEPT the null.

Calculated slope and intercept

~B1 = 0.5136

~B0 = -1.0781