![]() ![]() ![]() You will note that new stats sheets and new model sheets are assigned default names in ascending order of names not already taken. Click the History button at any time for an annotated list of all sheets in the file, with details of models, from which you can jump directly to any one of them. Here's the fast way to run all the analyses by clicking the buttons, starting from the data worksheet: Click Right to go to Stats 1 sheet > click Descriptive Statistics > click Run and wait for creation of Stats 3 > click Last Stats to return to Stats 1 Click Right to go to Stats 2 sheet > click Descriptive Statistics > click Run and wait for creation of Stats 4 > click Last Stats TWICE to go to to Stats 2 and back to Stats 4 Click Right to go to Model 1.0 sheet > click Linear Regression > click Run and wait for creation of Model 1.1 > click Last Model to return to Model 1.0 Click Right to go to Model 2.0 sheet > click Linear Regression > click Run and wait for creation of Model 2.1 > click Last Model to return to Model 2.0 Click Right to go to Model 3.0 sheet > click Linear Regression > click Run and wait for creation of Model 3.1 > click Last Model to return to Model 3.0 Click the Data button at any time to return to this sheet, click the Summaries button at any time to go to the Model Summaries sheet. You'll find that this a more effficient way to navigate the large amount of output that is in the file. As you are working, use the buttons on RegressIt's ribbon to move around in the file and to move up and down on worksheets. A picture of the RegressIt ribbon is shown above. If you are not running RegressIt on your computer at the moment, just browse through the file or this document and look at the screen shots of dialog boxes as well as the output and comments in order to see what it can do. If you do encounter technical problems, go to the tech-support web page on the web site and follow the directions there. It should only take you a few minutes to re-run all the analyses if there are no technical problems. The analyses use most of the table and graph output options at one point or another. An identical output worksheet should be created, and you can hit the Last Stats or Last Model button to toggle back and forth to the original for comparison. For purposes of testing RegressIt on your computer, you just need to download the file from the Downloads web page and go to each of the analysis worksheets in it and hit either the Descriptive Statistics button or the Linear Regression button according to the sheet type, followed by Run, to verify that the various analysis options are all working. Two models have been fitted in this file, having 1 and 8 variables, and descriptive analysis has also been performed. The objective is to build a model for predicting fuel economy from the other variables, which include weight, horsepower, displacement, acceleration, cylinders, and country of origin. It can be found on Kaggle and many other web sites.) There are 392 complete rows of data containing information for makes and models of cars sold in the U.S. (See the last worksheet in the file for documentation and sources. St.AutoData test file for RegressIt ( ): instructions and screen shots copied from Excel file This is a well-known data set that is widely used to demonstrate regression software and machine learning. Fabulous for the price but is a “point and shoot” kind of product rather than the “do all, show all” statistical tool one might expect for 25 times the price. Thus variational studies based on simulated data sets can not be accomplished by use of a single tool but would require use in concert with a spreadsheet.Ĭonclusion: A very useful and intuitive tool. What it Does Not Do: This is not a spreadsheet tool. The Gaussian application is interesting as it provides an easy way to establish a Gaussian curve based on left tail data as might be used in a reliability study of early failure data. This program handles more common math equations and also includes the Gaussian distribution. Analysis can be repeated for other equation forms based on original data set. This is useful because curve fits tend to fit the middle but not the ends of data sets - such behavior can be easily identified. Graphing of result enables comprehension of curve fit trends relative to data. Intuitive interface enables data input and computation of corresponding equation with ease. What it Does: Straight forward regression (Curve Fit) program for Mac OSX. Determines Equation that Fits Data & Graphs Result ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |