The Yearly Time-Series Wizard is a tool that helps you construct the various time-series expressions supported by WEAP's Data View. These expressions include functions for interpolation, step functions, smooth curves and linear, exponential and logistic projections. (You may also import many data expressions at once from Excel. See Export to Excel, Import from Excel for details.)
To access the wizard, either right-click on the data table or click on the down arrow on the right side of the expression box, and choose Yearly Time-Series Wizard from the menu.
The wizard is divided into three pages, which you step through using
the Next (
) and
Previous (
) buttons.
Use this page to select the type of function you want to create. The functions are summarized in graph form on screen as shown below, and are grouped into two main types.
The three functions on the top row allow you to specify data points for various future years, and the function then calculates the values for intervening years:
The interpolation function calculates values based on a linear (straight line) interpolation between the values you specify.
The step function assumes that values change discretely at the specified data years. In other words, values stay constant after a specified data year, until the next specified data year.
The smooth curve function calculates a best-fit smooth curve based on a polynomial least-squares fit of the specified data points. To achieve a good fit, the smooth curve function requires at least 3 data points.
The interpolation and smooth curve functions are most useful when you expect data to change gradually (for example when modeling the gradual penetration of some common device such as refrigerators or vehicles). The step function is most useful for specifying "lumpy" changes to your system, such as the construction of new transmission links.
The last three functions allow you to specify historic data values (i.e., values before the Current Accounts). The different functions are then used to extrapolate data forward to calculate future values. Extrapolations are based on linear, exponential or logistic least-squares curve fits. Use these functions with care. The onus is on you to ensure that the projections are reasonable, both in terms of how a) well the estimated curve fits the historical data, and b) how policies and other structural factors may change in the future. In other words, be sure to consider how well you can identify past trends, but also if it is reasonable to expect these past trends to continue into the future. WEAP helps you with task a) by providing various statistics describing the curve-fit: the R2 value, the standard error, and the number of observations. If you need to do a more detailed analysis, we suggest you use the data analysis features built-in to Microsoft Excel, and then link your results to your WEAP analysis (see below).
On page 2, you select the source of the data for the expression. Select whether you want to enter the data directly (i.e., type it in) or whether you want to link to the values in an external Excel spreadsheet.
Depending on your choice on page 2, in page 3 you either enter the data used by the function or select an Excel spreadsheet and range from which to extract the data for the selected time-series function.
When entering data directly, use the Add
(
) and Delete (
) buttons to add or delete new year/value pairs, or click
and drag data points on the adjoining graph to enter values graphically.
For the Interpolation function, an additional data field is shown allowing
you to specify a percentage growth rate, which is applied after the last
specified data year. By default this value is zero. In other words, by
default values are not extrapolated past the last interpolation data year.
The data you entered will be shown as the points on the adjoining chart,
while the line drawn on the chart will reflect the projection method you
chose on page one.
When linking to a Microsoft Excel sheet, first enter the name of the worksheet file (.XLS or .XLW) or use "..." button to browse your PC and local area network for the file. Next enter the range name from which the data will be extracted, or click the button attached to the field to select from the named ranges in the worksheet. Ranges can be specified either as names, or as Excel range formulae (e.g. Sheet1!A1:B16). NB: ranges must contain only two columns of data. The first column must contain years, arranged in chronological order (with the earliest at the top), and the second must contain data values. Click on the Get Excel Data button to extract the data from Excel and preview the values in the adjoining graph. Notice that the points on the chart will be the values in the Excel spreadsheet, while the line drawn on the chart will reflect the projection method you chose on page one.
See also: Data View, Expressions, Examples of Expressions, Export to Excel, Import from Excel, ExpForecast, Interp, LinForecast, LogisticForecast, Smooth, Step