{"id":2243,"date":"2018-09-19T07:32:31","date_gmt":"2018-09-19T12:32:31","guid":{"rendered":"http:\/\/variation.com\/?page_id=2243"},"modified":"2018-09-28T21:21:13","modified_gmt":"2018-09-29T02:21:13","slug":"change-point-analyzer-tutorial-1","status":"publish","type":"page","link":"https:\/\/variation.com\/change-point-analyzer-tutorial-1\/","title":{"rendered":"Change Point Analyzer – Tutorial 1"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Home<\/a>\u00a0\u00bb\u00a0<\/span>Software<\/a>\u00a0\u00bb\u00a0<\/span>Change-Point Analyzer<\/a>\u00a0\u00bb Tutorial 1<\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t

Change Point Analyzer Tutorial 1 - Getting Started<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Introduction<\/h3>

Before starting this tutorial, install Change-Point Analyzer on your computer.\u00a0 This tutorial will walk you through the steps of using this software to analyze a simple data set.\u00a0 We will be using the trade deficit data that appears in Donald Wheeler\u2019s book:\u00a0Understanding Variation \u2013 The Key to Managing Chaos.\u00a0 Table 1 shows the\u00a0US\u00a0trade deficit each month for 1987 and 1988 in billions of dollars.<\/p>

Table 1:\u00a0 US Trade Deficits 1987-1988 ($ billions)<\/strong><\/p>

<\/div>
<\/th>Jan <\/th>Feb <\/th>Mar <\/th>Apr <\/th>May <\/th>Jun <\/th>Jul <\/th>Aug <\/th>Sep <\/th>Oct <\/th>Nov <\/th>Dec <\/th><\/tr><\/thead>
1987 <\/td>10.7 <\/td>13.0 <\/td>11.4 <\/td>11.5 <\/td>12.5 <\/td>14.1 <\/td>14.8 <\/td>14.1 <\/td>12.6 <\/td>16.0 <\/td>11.7 <\/td>10.6 <\/td><\/tr>
1988 <\/td>10.0 <\/td>11.4 <\/td>7.9 <\/td>9.5 <\/td>8.0 <\/td>11.8 <\/td>10.5 <\/td>11.2 <\/td>9.2 <\/td>10.1 <\/td>10.4 <\/td>10.5 <\/td><\/tr><\/tbody><\/table><\/div>WP Table Builder<\/a><\/p>

Step 1:\u00a0 Start the Program<\/h3>

Start Change-Point Analyzer.\u00a0 Your screen should appear as in Figure 1.\u00a0 The blank Data window is used to enter the data.<\/p>

\"\"<\/p>

Figure 1:\u00a0 New Session<\/strong><\/p>

Step 2:\u00a0 Enter the Data and Labels<\/h3>

The data is generally entered in columns.\u00a0 You can also enter labels for the data points.\u00a0 We will enter both labels and data as shown in Figure 2.\u00a0 Start entering the labels by clicking on cell A1 (Column A and Row 1).\u00a0 Cell A1 will be highlighted in yellow.\u00a0 Type the word \u201cMonth\u201d and press the Enter key twice.\u00a0 This causes cell A3 to be highlighted.\u00a0 Type the first label, \u201cJan \u201887\u201d.\u00a0 Continue typing the rest of the labels, pressing the Enter key after each label to move to the next cell.\u00a0 Next enter the data.\u00a0 Start by typing \u201cTrade Deficit\u201d into cell B1.\u00a0 Use the scroll bars to move to this cell.\u00a0 Move to cell B3 and start entering the data.<\/p>

\"\"<\/p>

Figure 2:\u00a0 Data Window with Data Entered<\/strong><\/p>

Once the data has been entered, you can print a copy of the data for your records by clicking on the Print button.\u00a0 You can also paste data into the Data window using the clipboard.\u00a0 This allows you to easily analyze data you have already entered elsewhere.\u00a0 Change-Point Analyzer also comes with an Add-In for Excel that, once installed, allows you to initiate the analysis from within Excel.\u00a0 Tutorial 3 describes how to use this Add-In.\u00a0 The Trade Deficit data contains a single observation per time period.\u00a0 Change-Point Analyzer can also handle datasets containing multiple observations per time period.\u00a0 Tutorial 2 describes how to handle this type of data.<\/p>

Step 3:\u00a0 Select the Label Column<\/h3>

To select column A as labels, click the header for column A.\u00a0\u00a0 This will highlight the entire column.\u00a0 Then click the Label button.\u00a0 The label icon will appear in the column header indicating the column has been selected.\u00a0 Labels are optional.\u00a0 If no labels are specified, the row numbers are used as labels.<\/p>

Step 4:\u00a0 Perform the Analysis<\/h3>

Before analyzing the data, you must first select the data you want to analyze.\u00a0 To do this, click on the header for column B.\u00a0 This will highlight the entire column.\u00a0 Next, click the Fast Analysis button\u00a0\"\"\u00a0at the bottom of the Data window.\u00a0 Now sit back as the analysis is performed.<\/p>

As the analysis proceeds, the Analysis Status dialog box shown in Figure 3 is displayed.\u00a0 The purpose of this dialog box is to assure you that the analysis is progressing.\u00a0 Don\u2019t worry about understanding the status messages.\u00a0 Anything of importance will be displayed in the Analysis Results window once the analysis is completed.\u00a0 Analyses take from less than a second to several minutes depending on the speed of your machine and the amount of data you are analyzing.\u00a0 For the trade deficit data, it should only take 1-2 seconds to complete the analysis.<\/p>

\"\"
Figure 3:\u00a0 Analysis Status Dialog Box<\/strong><\/p>

Change-Point Analyzer automatically checks for outliers and a violation of assumptions.\u00a0 If either is detected, a message appears describing the problem and recommending how to proceed.\u00a0 A custom analysis is required.\u00a0 Tutorial 4 describes the general procedure for performing custom analyses while Tutorials 5 and 6 describe how to handle outliers and a violation of assumptions.<\/p>

Step 5:\u00a0 Interpret the Results<\/h3>

Once the analysis has been completed, the results are shown in the Analysis Results window in Figure 4.\u00a0 The results are displayed on seven tabs.\u00a0 We will examine each of these seven tabs.<\/p>

Step 5.1:\u00a0 Plot – Values Tab<\/h3>

The first tab shows a plot of the data (wavy black line).\u00a0 It also summarizes the results of the change point analysis in blue.\u00a0 The blue region shifts twice, once around June \u201887 and again around Nov \u201887.\u00a0 This indicates that the change-point analysis detected two changes in the trade deficit data.\u00a0 The second tab will provide further details.<\/p>

Also shown are two red lines, which are control limits.\u00a0 They represent the maximum range that the values are expected to vary over assuming no change has occurred.\u00a0 Points outside the control limits indicate a change has occurred.\u00a0\u00a0\u00a0\u00a0These control limits assume the values come from the normal distribution and may not be appropriate for all sets of data.\u00a0\u00a0The fact that point 12 is above the upper control limit also indicates that some sort of change occurred.\u00a0 Control charting is an alternative approach to detecting changes.\u00a0 As we will see, a change-point analysis has many advantages over control charting and is the preferred approach when analyzing historical data.\u00a0 However, Change-Point Analyzer also provides control charts of the data.<\/p>

\"\"
Figure 4:\u00a0 Analysis Results Window Showing Tab 1<\/strong><\/p>

You can hide the control limits, not display the blue region, represent the data with points rather than a line, change the title, scale and colors and much more.\u00a0 To do so, right-click the mouse over the plot.\u00a0 You can also click the Menu button\u00a0\"\".\u00a0 This displays the popup menu in Figure 5 for making these changes.\u00a0 The other tabs also have popup menus associated with them for customizing their display.<\/p>

\"\"
Figure 5:\u00a0 Popup Menu for Modifying Plot<\/strong><\/p>

Take a minute to make some changes to your plot.\u00a0 When you are done, you can print it by clicking the Print button\u00a0\"\"\u00a0on the bottom of the Data Analysis window.\u00a0 You can also copy the plot to the clipboard by clicking the Copy button\u00a0\"\".\u00a0 This allows you to paste the plot into your word processor using the Paste menu item on its Edit menu.<\/p>

Step 5.2:\u00a0 Table Changes – Values Tab<\/h3>

The second tab shows the results of the change-point analysis in table form (Figure 6).\u00a0 Each change detected is listed along with further information describing the change.<\/p>

\"\"
Figure 6:\u00a0 Analysis Results Window Showing Tab 2<\/strong><\/p>

The analysis detects two changes.\u00a0 The first change is estimated to have occurred around Jun \u201887.\u00a0 This point represents the first month following the change.\u00a0 The second change is estimated to have occurred around Nov \u201887.\u00a0\u00a0 Associated with each change is a confidence level indicating how confident the analysis is that the change actually occurred.\u00a0 The first change occurred with 90% confidence.\u00a0 The second change occurred with 100% confidence.\u00a0 We are much more confident about the second change.<\/p>

Also associated with each change is a confidence interval for the time of the change indicating how well the time of the change has been pinpointed.\u00a0 95% confidence is used for all confidence intervals.\u00a0 With 95% confidence, the first change occurred between May \u201887 and Jul \u201887.\u00a0 With 95% confidence, the second change occurred at Nov \u201887.\u00a0 The fact that the confidence interval for the first change is wider indicates that the time of the first change cannot be as accurately pinpointed as the second change.<\/p>

The second tab also gives additional information about each change.\u00a0 The table indicates that prior to the first change the average trade deficit was 11.82 billion dollars while after the first change it was 14.32 billion dollars.\u00a0 Tab 2 also gives a level associated with each change.\u00a0 The level is an indication of the importance of the change.\u00a0 The level 1 change is the first change detected and that which is most visibly apparent in the plot in Figure 4.\u00a0 Level 2 changes are detected on a second pass through the data.\u00a0 Any number of levels can exist depending on the number of changes found.<\/p>

Figure 4 represented these two changes by the shifts in the blue background.\u00a0 The blue background represents a region expected to contain all the values based on the current model that two changes occurred.\u00a0 Since all points fall within this region, this model fully explains the variation in the data.<\/p>

While the control chart in Figure 4 barely detected any change had occurred (one point barely out), the change-point analysis detected two changes.\u00a0 It also provided additional details including confidence levels and confidence intervals.\u00a0 This example illustrates two of the benefits of a change-point analysis: it is more powerful at detecting smaller sustained changes and it better characterizes such changes.\u00a0 When used to analyze historical data for trends and changes, a change-point analysis provides far more useful information than a control chart.\u00a0 For such data, the best approach is to perform a change-point analysis.\u00a0 However, this does not prevent one from also control charting the data.\u00a0 The results of both approaches can be displayed in a single plot as in Figure 4.<\/p>

Step 5.3:\u00a0 CUSUM – Values Tab<\/h3>

The third tab shows a cumulative sum chart (CUSUM) of the data (Figure 7).\u00a0 Change-Point Analyzer uses the CUSUM charts to identify the changes reported on Tab 2.\u00a0 You can ignore these plots if you desire.\u00a0 All the relevant results are given in Tab 2.<\/p>

Interpreting a CUSUM chart takes practice.\u00a0 Here are the basic rules:<\/p>