Learning Center

An essential step in an econometric analysis is a comprehensive investigation of the properties of your time series. Fortunately, the Econometrics Toolbox comes with a bundle of tools that provide standard econometrics routines that are necessary for a proper analysis of your data.

Maybe the best approach to introduce you with the Econometrics Toolbox data analysis capabilities is working with an example. In this section, we are going to analyze the following time series:

A

B

C

1

Sample  time series

 

 

2

 

 

 

3

Time

Value

 

4

2010-01-31

145

 

5

2010-02-28

150

 

6

2010-03-31

178

 

7

2010-04-30

163

 

8

2010-05-31

172

 

9

2010-06-30

178

 

10

2010-07-31

199

 

11

2010-08-31

199

 

12

2010-09-30

184

 

13

2010-10-31

162

 

14

2010-11-30

146

 

15

2010-12-31

166

 

16

2011-01-31

171

 

17

2011-02-28

180

 

18

2011-03-31

193

 

19

2011-04-30

181

 

20

2011-05-31

183

 

21

2011-06-30

218

 

22

2011-07-31

230

 

23

2011-08-31

242

 

24

2011-09-30

209

 

25

2011-10-31

191

 

26

2011-11-30

172

 

27

2011-12-31

194

 

28

2012-01-31

196

 

29

2012-02-29

196

 

30

2012-03-31

236

 

31

2012-04-30

235

 

32

2012-05-31

229

 

33

2012-06-30

243

 

34

2012-07-31

264

 

35

2012-08-31

272

 

36

2012-09-30

237

 

37

2012-10-31

211

 

38

2012-11-30

180

 

39

2012-12-31

201

 

40

2013-01-31

204

 

41

2013-02-28

188

 

42

2013-03-31

235

 

43

2013-04-30

227

 

44

2013-05-31

234

 

45

2013-06-30

264

 

46

2013-07-31

302

 

47

2013-08-31

293

 

48

2013-09-30

259

 

49

2013-10-31

229

 

50

2013-11-30

203

 

51

2013-12-31

229

 

52

2014-01-31

242

 

53

2014-02-28

233

 

54

2014-03-31

267

 

55

2014-04-30

269

 

56

2014-05-31

270

 

57

2014-06-30

315

 

58

2014-07-31

364

 

59

2014-08-31

347

 

60

2014-09-30

312

 

61

2014-10-31

274

 

62

2014-11-30

237

 

63

2014-12-31

278

 

64

2015-01-31

284

 

65

2015-02-28

277

 

66

2015-03-31

317

 

67

2015-04-30

313

 

68

2015-05-31

318

 

69

2015-06-30

374

 

70

2015-07-31

413

 

71

2015-08-31

405

 

72

2015-09-30

355

 

73

2015-10-31

306

 

74

2015-11-30

271

 

75

2015-12-31

306

 

To analyze this series, go the Econometrics Toolbox tab, select any cell within the series range and then click on the green arrow in the command window.

And that is it! Now you can check the following analysis of your data:

A series line chart (double click the chart to expand it):

A density plot:

The autocorrelation and partial autocorrelation chart:

A box plot to check seasonality:

Note: The default seasonality is set to 12. To adjust this, just right click your data and type the seasonal length that you wish. You can also define which series your series begin (for instance, if you are investigating seasons in months and your series begin in March, type 3 in the “first value season” field.

The summary statistics of your series:

You can go further in your analysis, analyzing any transformations of your data. Besides, if you are doing a regression analysis, you will have all the analysis mentioned above for the dependent variable (i.e., the series itself), the predicted series and the residuals.

You can also investigate just some portion of your series, using the “data input” selection tool. Subsetting your data is made by changing the sliders thumbs (or changing the text input values) to use only the fraction of your data that you wish. In addition, to check the effective range that will be used in the analysis, just click on the “Preview series” button located in the command window.

The Econometrics Toolbox also deals well with missing values in your series. Some valid characters to represent missing values is “*” and “NA”. Please, pay attention that an empty cell is NOT a valid missing value. The standard internal procedure is to exclude the entire row containing a missing data from analysis or to ignore the missing values when it is not used for any calculus, for instance, to plot the time series. In the latter case, SAFE TOOLBOXES® will not plot anything in the index of series where there is no value.

© 2016 Safe Quantitative Technologies, ltd. All rights reserved.