Perform simple time series analysis including autocorrelation, cross correlation, differencing, and single exponential smoothing.

INSTALLATION
Download the Simple Time Series Analysis.opx, and then drag-and-drop onto Origin workspace. An icon will appear in the Apps Gallery window.
NOTE: This tool requires OriginPro.

OPERATION
Click on the Simple Time Series Analysis icon to open toolbar with 5 buttons:

• Differences
  This tool carries out non-seasonal and seasonal differencing on time series data.
  1. Click DIFF button in the toolbar. A dialog will open.
  2. Choose XY data as Time Series.
  3. In the Settings branch, select Difference or Seasonal Difference as Type.
     When Type is Difference, check the checkbox on right of Difference Order and enter a positive integer.
     When Type is Seasonal Difference, seasonal and non-seasonal differences are allowed at the same time. If  non-seasonal difference is needed, check the checkbox on right of Difference Order and enter a positive integer. Otherwise clear the checkbox. In Seasonal Difference group, check the checkbox on right of Order. Enter a positive integer as Order and a positive integer as Period.
  4. Click OK button, the results of differences will be created.
• Lag
  This tool calculates lag or lead of a time series.
• Single Exponential Smoothing
  This tool performs single exponential smoothing on time series data.
  1. Click Single Exponential Smoothing button in the toolbar. A dialog will open.
  2. Choose XY data as Time Series.
  3. In the Settings branch, you can use default Smoothing Constant calculated by fitting an ARIMA (0, 1, 1) model or custom a value. The value must be between 0 and 1. The smaller the value, the smoother the curve. When you customize the Smoothing Constant value, Number of Points to Initialize should also be specified to estimate m0. Number of Forecasts is used to define how may forecasts you want. Zero means no forecast is required.
  4. In the Quantities branch, check options to control which results to output in Results sheet, including Residuals, Standard Error and Confidence Limits for forecasts.
  5. In the Plots tab, specify whether to create Smoothed Plot and Predicted Plot.
  6. Click OK button, a report sheet and a report data sheet will be created.
• Autocorrelations and Partial Autocorrelations
  This tool computes and plots autocorrelations and partial autocorrelations on time series data.
  1. Click ACF & PACF  button in the toolbar. A dialog will open.
  2. Choose XY data as Time Series.
  3. In the Settings branch, you can use default Number of Lags or specify a custom value. The default method used for computing standard error is Independence Model. You can change it to Bartlett's Approximation. Check the Partial Autocorrelation to output the corresponding results.
  4. In the Quantities branch, check options to control which results to output in Results sheet, including Standard Error, t Statistic, Confidence Limits, etc.
  5. In the Plots tab, specify whether to create a graph of autocorrelation and a graph of partial autocorrelation.
  6. Click OK button. A report sheet and a report data sheet will be created.
• Cross Correlation
  This tool computes and plots cross correlations between two time series data. The two data should have the same number of observations.
  1. Click CCF button in the toolbar. A dialog will open.
  2. Choose one Y dataset for Time Series 1 and Time series 2 respectively.
  3. You can use default Number of Lags or custom a value.
  4. In the Quantities branch, check options to control which results to output in Results sheet, including Standard Error and Confidence Limits.
  5. Click OK button, a report sheet and a report data sheet will be created.

v 1.2: bug fix
v 1.1: Fix bugs and improve CCF
v 1.02: Fix bug related to toolbar image