Demetra+ - Joinup



Demetra+

Prototype IV

[pic]

User Manual

Contents

1. Introduction 3

1.1. Overview of the software 3

1.2. Browsing the data 4

1.3. Importing data from Excel 5

1.4. Displaying the data 7

2. Application’s Menu 8

2.1. Workspace menu 8

2.1.1. Calendars 8

2.1.2. User-defined regression variables 9

2.2. Tools menu 11

2.2.1. Container 11

2.2.2. Spectral Analysis 13

2.2.3. Differencing 14

2.3. X-12 Doc 15

2.4. Window menu 19

2.5. Interaction with other tools 20

3. Single seasonal adjustment 21

4. Multi-processing 24

5. Changing the specification 16

6. Seasonal adjustment results 25

6.1. Main results 26

6.1.1. Pre-processing 30

6.1.2. Decomposition 35

6.1.3. Diagnostics 39

6.2. Editing and modifying the results 46

6.3. Sending the results to external devices 48

6.4. Saving and refreshing workspaces 49

Introduction

The aim of this short document is not to provide a complete description of the software but to give a first overview of the tool and of its main functionalities.

Demetra+ is designed as an open and flexible software, which allows various uses. Instead of describing the different functions and presentation tools of the application, we prefer, in this first document, to describe step by step the way to solve some very basic tasks.

It is clear that such a non systematic approach is not sufficient to understand completely the software. However, we think that it is the easiest and the most stimulating way to get some feeling of this new product.

In a first stage, we just consider the way to visualize the data provided with the software and to import new series from Excel.

In a second stage, we explain how we can adjust a single series.

In a third stage, we consider the processing of many series.

In a last stage, we discuss the way to save and to refresh a previous work.

1 Overview of the software

When it is opened for the first time, Demetra+ appears with the following layout:

[pic]

The key parts of the application are:

• the browsers panel (left panel), which presents the available time series

• the workspace panel (right panel), which shows information used or generated by the software

• a central blank zone that will contain actual analyses

• two auxiliary panels at the bottom of the application; the left, one (TSProperties) contains the current time series (from the browsers’ panel) and the right one (Logs) contains logging information.

Panels can be moved, resized, superposed, closed[1]... depending on needs or preferences of the user. The presentation is saved between different sessions of Demetra+.

The application can contain multiple documents. Following the needs, you can present them in different tabs taking the full space (default) or in floating windows (choose this one to follow different steps). The main menu item "Window->Floating/Tabbed..." gives access to that functionality.

2 Browsing the data

The browsers’ panel presents the series available in the software. Different "time series providers" are considered: Xml (specific schema), Excel, TSW, USCB, Text, ODBC...

The installation procedure has copied several files in different formats in the subfolders of "My Documents\Data". We explain below the way to open Excel workbooks. The procedure is similar for the other providers.

1. Click on the Excel tab of the browsers panel.

2. Click on the left button (see below).

3. Choose an Excel workbook (for instance "insee.xlsx") in the folder "My Documents\Data\Excel".

[pic]

Different browsers show the data in trees that can be expanded by double-clicking their nodes (or single-clicking the +/- signs)

[pic].

Final nodes of the trees represent time series and their parents represent collections of time series.

3 Importing data from Excel

Time series from Excel can easily be integrated in Demetra+. The series must be formatted in Excel as follows:

• True dates in the first copied column/row,

• Titles of the series in the first copied row/column,

• Empty top-left cell ,

• Empty cells in the data zone correspond to missing values (except at the beginning and at the end of the series).

That format corresponds with the format used by the Excel browser (which also requires that the input zone starts at the beginning of the sheet [A1]).

[pic]

After they have been copied in Excel, the data can be integrated in Demetra+ as follows:

• Select the xml panel in the browsers,

• Paste the data (they appear in the tree),

• Change the names of the series/collection in the tree if necessary (click twice on the item you would like to modify),

• Save the file (if need be).

[pic]

4 Displaying the data

When clicking on the one node of the trees, the basics statistics, chart and time series data are presented in TS Properties window automatically.

[pic]

1. Application’s Menu

1. Workspace menu

Workspace menu offers the following functions:

• New - creates new Workspace displayed in the right panel,

• Open - opens an existing project in a new window,

• Save - save the project file named by the system (workspace_#number) that can be re-opened at a later point in time,

• Save as - save the project file named by the user that can be re-opened at a later point in time,

• View - activates or deactivates the panels chosen by user (Browsers, Workspace, Logs, TS Properties)

• Edit - allows defining countries’ calendar and regression variables. This functionalities is described further into this instruction.

• Exit - closes an open project.

[pic]

1. Calendars

The calendars of Demetra+ simply correspond to the usual trading days contrasts variables, based on the Gregorian calendar, modified to take into account some specific holidays. Those holidays are handled as "Sundays" and the variables are properly adjusted to take into account long term mean effects.

Demetra+ considers three kinds of calendars:

• National calendars, identified by specific days

• Composite calendars, defined as weighted sum of other calendars

• Chained calendars, defined by two other calendars and a break date

The calendars can be defined recursively.

The data generated by each calendar can be viewed by a double click on the corresponding item in the workspace tree.

[pic]

The regression variables can be generated for any frequency and any (reasonable) time span through that window; the periodogram of those series are displayed when a column is selected.

Finally, the series can be copied by drag (see blue circle) and drop or through the local menu to other places/software

2. User-defined regression variables

User-defined regression variables are simply time series identified by their name. Those names will be used in other parts of the software (regression) as identifier of the data.

Demetra+ considers two kinds of user-defined regression variables:

• Static variables, usually imported from external software

• Dynamic variables, coming from some browser.

Static variables imported from external software (for instance Excel) must be formatted as defined in the "Quick Start" document. They are imported by drag and drop or using the usual keys (ctrl-v).

[pic]

Dynamic variables are imported by drag and drop from a browser of the application.

The figures of static variables cannot be changed. Currently, the only way to update static series consists in removing them from the list and to re-import them with the same names as previously.

Dynamic variables are automatically updated each time the application is re-opened. So, they should be the preferred way to handle user variables.

The names of the series can be changed as usual (selected a series and click once again when it has been selected). The selected series can be showed in a small chart window by a double click.

[pic]

2. Tools menu

1. Container

To display the data, it is also possible to open one of the available containers of Demetra+ (charts, grids or tables, lists...)

[pic]

The user can take any series or group of series from one of the browsers and drop it in a container.

[pic]

The group cannot be marked using Ctrl button. Add the series to chart or grid, by dragging and dropping them one by one.

The chart is automatically rescaled after adding new series. Also new item “Chart” is added to menu toolbar.

[pic]

Putting numerous time series into one chart could make it confusing. In this case it is helpful to click on one series which is then displayed in bold.

[pic]

The right-button menu offers many useful options. Chart can be edited, printed or saved as bitmap.

[pic]

2. Spectral Analysis

Tools menu offers spectral estimators – periodogram and autoregressive spectral estimator. After choosing it from Tools menu the empty window is displayed.

[pic]

To calculate periodogram drag and drop a raw time series into the displayed window. The methodological note about spectral analysis is available further into this instruction.

[pic]

3. Differencing

Using Differencing window from Tools menu it is possible to obtain not only table and spectral graphs but also ACF and PACF functions for selected time series. To do it, the time series’ name from the list should be dragged and dropped into Name box.

[pic]

Autocorrelation function (AC) displayed graphically and numerically in the graph, is the serial correlation coefficients of the current value of the series with the series lagged a certain number of periods in a range of lags (1 through 36). Green dotted lines indicate range of two standard errors.

Drag and drop is omnipresent in Demetra+. It is the usual way to move information between different components. The objects that can be moved (time series, collections of time series...) can take different forms: nodes in trees, labels in lists, headers in tables, lines in charts...

When a drag and drop operation is initiated (which means that an object is indeed "moveable"), the cursor of the mouse changes to either a "no parking sign" or to a "+ sign", which indicates an acceptable drop zone.

Information can also be moved by drag and drop or copy/paste to external software (for instance Excel).

3. X-12 Doc

This item is added to the application’s menu when X-12 seasonal adjustment was done ant it is active.

[pic]

Current specification is displayed in a non modal dialog box, so the user can change any option and inspect its impact on the results. For a detailed description of the X12 specifications, you should refer to the "Demetra_Spec.docx" document.

For example:

• Select in the output window the 'Main results -> Charts" window,

• Modify the span of the series in the "Basic" panel:

o Click on the Basic item in the left panel of the specification dialog box,

o Expand the "series span" node in the right panel,

o Choose the "excluding" selection type,

o Write "12" in the "last" node,

• Press the "Apply" button.

The processing is computed on the series without the last 12 observations. A visual comparison of the forecasts of X12 and of the actual figures is displayed on the chart.

[pic]

Some other examples are explained below. They are based on series coming from the "prod.xml" file, which provide more eloquent results. The next snapshots use the series named "Industries manufacturières"

Suppressing of trading days

The trading days regression variables can be suppressed by setting the "Trading days -> Type" to "None" in the "Calendar effects" panel of the specification dialog box.

[pic]

Meaningful information is provided in the "Pre-processing -> Arima" panel or in the different panels of the spectral analysis

[pic]

Changing the specification

The user is able to modify the used specification and to see immediately the result of changes made.

The specification is edited through the main menu: TramoSeatsDocxxx / X12Docxxx -> Specification... It is possible to edit the specification used to generate the processing (current specification) or the specification that corresponds to the results (result specification).

[pic]

Changing X11 options

The X11 panel of the specification dialog box contains a rich set of options on the X11 decomposition. Their effects appear - for instance - in the SI-ratio chart

[pic]

The previous snapshot was realized by setting the "Use forecasts" option on false and the "Seasonal filter" on "S3x15"

User-defined regression variables

Once calendar variables and/or user-defined regression variables have been defined (see general features), they can be integrated in the regression model of X12 using several ways.

Calendars can be chosen in the trading days node of the calendar effects panel of the specification dialog box:

• Select the option "Calendar" in the "Type" field,

• Choose the holiday in the list that is automatically displayed (it corresponds to the calendars defined in the workspace; calendars must be defined to get access to that option),

• The corresponding calendar variables are computed "on the fly".

[pic]

Trading days effects can also be defined in a free way as pure user-defined regression variables:

• Select the option "UserDefined" in the "Type" field,

• Choose the variables in the list that is displayed after a click on the dots on the right of the "Details -> Items" row (it corresponds to the user variables defined in the workspace; user variables must be defined to get access to that option).

[pic]

Regression variables that are not considered as trading days can be integrated in a quite similar way through the regression panel of the specification dialog box.

4. Window menu

Window menu offers the following functions:

• Floating,

• Tabbed - arranges all windows in central zone as tabs,

• Tile vertically -arranges all windows in central zone vertically,

• Tile horizontally -arranges all windows in central zone vertically,

• Skinning – allows to custom graphical appearance of Demetra+.

[pic]

The window menu includes also the seasonal adjustment processing done and not closed by the user during the current session.

“Documents” option offers some additional options helpful for organising windows.

[pic]

5. Interaction with other tools

The different series that appear in the results can be dropped in other windows of Demetra+

Useful examples are described below:

• Open auto-regressive spectrum windows using the following main menu item: "Tools -> Spectral Analysis -> Auto-regressive spectrum",

• Drop meaningful series of the results in those tool windows (for example "A1" and "B1" from the "Decomposition (X11)" sub-nodes,

• Open a "growth chart" using the following main menu item: "Tools -> Tool Windows -> Growth chart",

• Drop D11 (for example) in it.

[pic]

The different tool windows are dynamically updated when:

• A new series is selected, through a double click in the browsers panel or when a series is dropped in the left zone of the X12 window,

• The specification is changed, by means of the specification dialog box or when another specification, coming from the workspace, is dropped in the left zone of the X12 window.

Many other combinations are of course possible.

It should be noted that the current implementation is not able to detect recursive processing, obtained for instance by dropping a series like "D11" in the left zone of the same X12 window. Such an attempt will generate a crash of Demetra+.

2. Single seasonal adjustment

The seasonal adjustment of a single series can be done in different ways. Below, we present the simplest solution (also known as the short –cut).

The first step to produce a fast seasonal adjustment consists in choosing a specification from the list displayed in the workspace panel. The procedure is as follows:

• Select in the tree the specification you want to select,

• Open the local menu by means of the right button of the mouse,

• Choose the "Active" menu item.

[pic]

That specification, called active specification, will be used to generate the processing. It can be changed at any time.

Specifications correspond to the terminology used in TSW[2].

|Name |Content |

|RSA1 |log/level, outliers detection, Airline model |

|RSA2 |log/level, working days, Easter effect, outliers detection, Airline model |

|RSA3 |log/level, outliers detection, automatic model identification |

|RSA4 |log/level, working days, Leap year, Easter effect , outliers detection, automatic model identification |

|RSA5 |log/level, trading days, Leap year, Easter effect, outliers detection, automatic model identification |

Once the active specification is chosen, the user has just to make a double click on the series in the browsers’ panel that he wants to adjust. The processing is immediately initiated, with the selected specification and the chosen series.

[pic]

Demetra+ provides a multiple documents interface (MDI). The user can display and compare many documents.

One can process as follows:

• Select another active specification (for instance, "RSA1"), following the procedure described above,

• Double click on a series in the browsers panel,

• Make some place by clicking on either the "Auto Hide" and "Close" buttons of the system windows ("Workspace", "Logs", "TSProperties"),

• Tile the output windows through the "Window" sub-menu,

• New double click in the browsers will update the different processing (that feature can be disabled by locking a given processing, through the "X12Doc-xxx" -> "Lock" main menu item.

•

This procedure can be used with any specification (Tramo-Seats and X12).

The figure below shows how to use the "Auto Hide" (red) button or the "Close" (blue) button:

[pic]

As an example, the following chart presents the comparison of the results for “Tile vertically” option.

[pic]

The user can inspect the different facets of the results through the exploring tree displayed in the left panel of the output window. The results contain many detailed panels. The user can go through them by selecting a node in the navigation tree of the X12 processing. The current specification and the current series are displayed on the top of the window.

3. Multi-processing

The seasonal adjustment of many series can also be done using various ways. The short way is explained below.

Multi-processing needs the following steps:

• Create a new multi-processing window, through the main menu (Seasonal adjustment -> Multi-processing -> New),

• Choose an active specification (see above),

• Import by drag and drop the series (or collection of series) you want to process; they can come from different providers.

If you want to mix different specifications (from TramoSeats and X12), you can change at any time the active specification; the series added after that change will be associated with the new active specification.

Start the processing (through the main menu "SAProcessing XXX-> Start"); it should be noted that the processing runs in a multi-threaded environment; so, the user can continue his work without waiting for the end of the action.

[pic]

4. Seasonal adjustment results

The exact content of the panel depends on the seasonal adjustment methods chosen by the user. The basic output structure is as follows:

• Main results,

o Charts,

o Table,

o S-I ratio,

• Pre-processing,

o Pre-adjustment series,

o Arima,

o Residulas,

• Decomposition,

o Stochastic series,

o WK analysis,

• Diagnostics,

o Seasonality tests,

o Spectral analysis,

o Revision history,

o Sliding spans,

o Model stability.

Detailed description of the seasonal adjustment outcomes is presented below.

1. Main results

This section includes basic information about seasonal adjustment and the quality of the outcomes. Left panel presents the results from Tramo Seats, right panel - the results from X-12-ARIMA

[pic]

In Charts section the top panel presents the original series with forecasts, the final seasonally adjusted series, the final trend with forecasts and the final seasonal component with forecasts. The second panel shows the final irregular component and the final seasonal component with forecasts.

[pic]

The charts in main results section (and many other charts available in Demetra+) provide zooming features. Select the zone you want to zoom with your mouse (from the top-left corner to the bottom-right one). Select any rectangle from bottom-right to top-left to un-zoom the chart.

[pic]

You can move (red) or resize (blue) the selected zooming window with the scroll bars. You can also highlight any series by clicking on it.

[pic]

Table presents the original series with forecasts and forecast error, the final seasonally adjusted series, the final trend with forecasts, the final seasonal component with forecasts, the final irregular component.

[pic]

S-I ratio chart presents the final estimation of the seasonal-irregular component and final seasonal factors for each of the period in time series (months or quarters). The combined seasonal-irregular component (dots) is obtained by subtracting trend-cycle from Original Series Adjusted for Trading-day and Prior Variation (additive model) or division the trend-cycle by Original Series Adjusted for Trading-day and Prior Variation (multiplicative model). Curves represents the final seasonal factors and the straight line represents the average for the these values in each period.

[pic]

You can enlarge a specific period in the SI-ratio chart by clicking in its zone. The details are displayed in a resizable pop-up window.

[pic]

1. Pre-processing

Pre-adjustment series

Table presented in pre-processing section contains series estimated in TRAMO or Reg-ARIMA part. It includes interpolated series, series adjusted for calendar effects, deterministic component, calendar effects, trading days effect, outliers effect on irregular component, total outliers effect, total regression effect (X-12ARIMA only).

Arima

This section demonstrates spectrum linearised series and ARIMA model parameters’ values.

[pic]

Residuals

Residuals are calculated as a difference between the one-step-predicted output from the model and the measured output from the validation data set. They represent the portion of the validation data which is not explained by the model. One-step ahead residuals from the model are presented in the graph and in the numbers.

[pic]

Analysis of the residuals consist of several tests.

Normality test

Jarque-Berra test on normality checks if skewness and kurtosis in the sample is like in the normal distribution. The assumption of normality is used to calculate the confidence intervals from standard errors. If this assumption is rejected, the estimated confidence intervals will be distorted even if the standard forecast errors are reliable. A significant value of one of these statistics indicates that the standardised residuals do not follow a standard normal distribution; because of that the reported confidence intervals might not represent the actual ones.

[pic]

Independence of the residuals

This statistic measures the probability of the autocorrelation function occurring under the hypothesis that the model has accounted for all serial correlation in the series up to the lag. P-values greater than 0.05, up to lag 12 for quarterly data and lag 24 for monthly data, indicate that there is no significant residual autocorrelation and that, therefore, the model is adequate for forecasting.

[pic]

Durbin-Watson test

The Durbin-Watson statistic is a test for first-order serial correlation. This statistic verified the linear association between residuals from a regression model, testing if [pic]in the following specification:

[pic],

Where:

[pic]- residuals.

If there is no serial correlation, the DW statistic will be around 2. The Durbin-Watson statistic ranges in value from 0 to 4. If there is no serial correlation, the DW statistic will be near 2. The values lower than 2 indicates positive autocorrelation, while the values higher than 2 indicates negative autocorrelation.

Randomness of residuals

The runs test implemented in Demetra+ is used to test for randomness of residuals. To conduct the test a up-down series is created. The observations in this series indicate if the given value in is up or down in value from the prior value in sequence. The runs test is performed on the up-down series rather than upon the original quantitative variable.

[pic]

Linearity of the residuals

Residual autocorrelation is verified using two tests: Ljung-Box test and Box Pierce test.

For Ljung –Box the test statistic is:

[pic],

where:

[pic] - the sample size,

[pic] - the squared sample autocorrelation at lag  [pic]

[pic] - the number of lags being tested.

Null hypothesis is: The data are random.

The test statistic for Box-Pierce test is:

[pic]

where:

[pic]- [pic]sample autocorrelation,

[pic]

The p-value marked in red indicates that null hypothesis was rejected.

Demetra+ also presents residuals’ distribution. In this section autocorrelation and partial autocorrelation functions are presented.

[pic]

2. Decomposition

The contain of Decomposition section depends on the seasonal adjustment method chosen by user.

Tramo/Seats

The decomposition made by Seats assumes that all components in time series - trend, seasonal and irregular - are orthogonal and could be modeled using ARIMA model. Identification of the components requires that only irregular components include noise. ARIMA models estimated for each components are presented below:

[pic]

Additional information presented by Demetra is set of stochastic series (seasonally adjusted series, trend, seasonal component, irregular component, trend-forecast, seasonal component-forecast) and Wiener-Kolmogorow analysis.

X-12-ARIMA

In the decomposition section all important tables from X-11 procedure are available:

Part A - Raw Series

Table A1- Time Series Data

Table A1a -

Table A2 -

Table A6 -

Table A8 -

Part B - Preliminary Estimation of Extreme Values and Calendar Effects

Table B1 - Raw Series Adjusted a Priori

Table B2 - Trend-Cycle

Table B3 - Unmodified Seasonal-Irregular Component

Table B4 - Replacement Values for Extreme SI Values

Table B5 - Seasonal Component

Table B6 - Seasonally Adjusted Series

Table B7 - Trend-Cycle

Table B8 - Unmodified Seasonal-Irregular Component

Table B9 - Replacement Values for Extreme SI Values

Table B10 - Seasonal Component

Table B11 - Seasonally Adjusted Series

Table B13 - Irregular Component

Table B17 - Preliminary Weights for The Irregular

Table B20 - Adjustment Values for Extreme Irregulars

Part C - Final Estimation Of Extreme Values And Calendar Effects

Table C1 - Modified Raw Series

Table C2 -Trend-Cycle

Table C4 - Modified SI

Table C5 - Seasonal Component

Table C6 - Seasonally Adjusted Series

Table C7 - Trend-Cycle

Table C9 - SI Component

Table C10 - Seasonal Component

Table C11 - Seasonally Adjusted Series

Table C13 - Irregular Component

Table C20 - Adjustment Values For Extreme Irregulars

Part D - Final Estimation of the Different Components

Table D1 - Modified Raw Series

Table D2 -Trend-Cycle

Table D4 - Modified SI

Table D5 - Seasonal Component

Table D6 - Seasonally Adjusted Series

Table D7 - Trend-Cycle

Table D8 - Unmodified SI Component

Table D9 - Replacement Values for Extreme SI Values

Table D10 - Final Seasonal Factors

Table D10A -

Table D11 - Final Seasonally Adjusted Series

Table D11A - Final Seasonally Adjusted Series with Revised Annual Totals

Table D12 - Final Trend-Cycle

Table D12A -

Table D13 - Final Irregular Component

Table D13U -

Table D16 - Seasonal and Calendar Effects

Table D16A -

Table D18 - Combined Calendar Effects Factors

Part E - Components Modified For Large Extreme Values,

Table E1 - Raw Series Modified for Large Extreme Values

Table D2 - SA Series Modified for Large Extreme Values

Table E3 - Final Irregular Component Adjusted for Large Extreme Values

Table E11 - Robust Estimation of the Final SA Series

[pic]

3. Diagnostics

Seasonality tests

Several seasonality tests are available in Demerta+.

Friedman test

Friedman's test is a non-parametric method for testing that samples are drawn from the same population or from populations with equal medians. In the regression equation the significance of the month (or quarter) effect is tested. Friedman test requires no distributional assumptions. It uses the rankings of the observations. Therefore it is not as sensitive to outliers in small samples as regression analysis. The test statistic is approximately distributed as chi-squared with 11 degrees of freedom. The null hypothesis is: The months (quarters) have identical effects.

Kruskal-Wallis test

Kruscal-Wallis test is used to compare samples from two or more groups. The null hypothesis states that all months (or quarters, respectively) have the same mean.

[pic]

Spectral analysis

Demetra+ provides spectral plots to alert the user to the presence of remaining seasonal and trading day effects. Two spectrum estimators are implemented.

Periodogram has the following form[3]:

[pic]

Where:

[pic]- frequency, [pic]

[pic]- time series, [pic].

The second spectrum estimator is an autoregressive spectrum estimator[4]:

[pic]

Where:

[pic]- frequency, [pic]

[pic]- the sample variance of the residuals

[pic]- coefficients from regression [pic] on [pic], [pic].

Seasonal frequencies are marked as grey, vertical lines, while violet lines correspond to trading-days frequencies. The X-axis shows the different frequencies. The periodicity of phenomenon at frequency[pic] is[pic]. It means that for monthly time series the seasonal frequencies are: [pic]. Horizontal yellow line corresponds to a significance level of 0.5% and delimits highly suspect peaks.

[pic]

At seasonal and trading days frequencies, a peak in model residuals indicates that the need for a better fitting model. A peak in the spectrum from the seasonally adjusted series or irregulars reveals inadequacy of the seasonal adjustment filters for the time interval used for spectrum estimation. In this case different model specification or data span length should be considered.

Revision histories

Revision histories deal with the revisions associated with continuous seasonal adjustment over a period of years. Revisions are calculated as a difference between the first (earliest) adjustment of an observation computed when that observation is the final period of the time series (concurrent adjustment, denotes as [pic]) and a later adjustment based on all data span (most recent adjustment, denotes as [pic]).

The revision history of the seasonal adjustment from time [pic]to [pic] is a sequence of [pic] calculated in a following way[5]:

[pic]

Demetra+ calculates revision history for both SA series and trend-cycle component. Revision history is done for last 12 (monthly series) or 4 (quarterly) observations.

Red line represents the final estimation of the seasonally adjusted time series, while dots denote the initial estimation. The history analysis plot is accompanied by information about the relative difference between initial and final estimation. Values which absolute value are too large are marked in red and provide information about the stability of the outcome.

[pic]

In the revisions history panels the user can have a complete overview of the different revisions for a given time span by selecting with the mouse (just like for zooming) the considered periods. The successive estimation are displayed in a separate pop-up window

[pic]

One can also get all the revisions for a specific period by clicking on the point that corresponds to the first estimate for that period. The results of those popup-up windows can be copied or drag and dropped to other software (Excel...)

[pic]

By default, the revisions are obtained with the same model, but in re-estimating its parameters. That option can be changed through the local menu of the revision history node (left panel), at the expense of the speed of the processing and for results that are usually very similar.

Sliding spans

The sliding spans analysis means the comparison of the correlated seasonal adjustments of a given observation obtained by applying the adjustment procedure to a sequence of three or four overlapping spans of data, all of which contain this observation. The procedure of withdrawing spans from time series is described in Findley, Monsell, Shulman and Pugh (1990)[6]. Each period (month or quarter) which belongs to more than one span is examined to see if its seasonal adjustments vary more than a specified amount across the spans.

Seasonal factor is regarded to be unreliable if the following condition is fulfilled:

[pic], (16)

Where:

[pic]- the seasonal factor estimated from span [pic] for month [pic].

[pic] = {[pic]: period [pic] is in the [pic]-th span}; .

For seasonally and trading days adjusted series the following statistic is being calculated:

[pic]

The value is considered to be unreliable if it is higher than [pic].

Similarly, the seasonally adjusted changes are unstable if :

[pic]

Where:

[pic]- the seasonally (or trading day) adjusted value from span [pic] for month [pic]

[pic]= {[pic]: period [pic] and [pic] are in the [pic]-th span} .

Results of sliding spans analysis are presented in three graphs. Upper panel shows the sliding spans statistic for each period, the bottom-left panel presents the distribution of sliding spans statistics and bottom-left panel contains information about the percentage of values for which sliding spans condition is not fulfilled.

[pic]

One can also change the series and/or the specification by importing new elements[7] by drag and drop (in the exploring tree).

Model stability

The diagnostics output window provides some purely descriptive features to analyze the stability of some part of the model, like trading days, Easter and Arima. For instance, the coefficients of the trading days variables computed by means of a moving window (by default on 8 years, using the same model but with re-estimated parameters) are displayed in the panel corresponding to the "Model stability -> Trading days" node.

[pic]

2. Editing and modifying the results

It is possible to edit any item of the processing by double clicking it.

The complete output, with the used specification is displayed. The user can modify it, apply the new specification to see its effects and save the results to the multi-processing window, if need be.

[pic]

It is not necessary to close the details window to get information on another series; that window is updated by a simple click on another series of the multi-processing view.

3. Changing the specification

The user is able to modify the used specification and to see immediately the result of changes made.

The specification is edited through the main menu: TramoSeatsDocxxx / X12Docxxx -> Specification... It is possible to edit the specification used to generate the processing (current specification) or the specification that corresponds to the results (result specification).

[pic]

4. Sending the results to external devices

When the processing is finished, it is possible to generate several outputs (Excel workbook, csv files...), through the main menu command: "SAProcessingXXX -> Generate output...". It should be noted that Excel and .csv outputs will be put in the temporary folder if their target folders are not specified

[pic]

5. Saving and refreshing workspaces

By default, single and multi-processing generated through the so-called "short-ways" are not put in the current workspace. To be able to save and to refresh them, the user must first add them to the workspace. That can be done, for instance, through the main menu "SAProcessingXXX -> Add to Workspace".

[pic]

The user still has to save the workspace, using the usual menu command (Save/Save as).

When Demetra+ is re-opened, it will automatically open at the last used workspace. The software also maintains a list of the most recently used workspace, which can be easily accessed.

[pic]

A saved item of a workspace can be opened by a double click or by its local menu. It is then showed in its previous state. Demetra+ proposes several options to refresh it[8]:

[pic]

|Parameters |Only the parameters are refreshed. The structure of the model is unchanged |

|Outliers (+ params) |Outliers (and of course parameters) are re-estimated |

|Arima and outliers (+params) |Stochastic component, outliers (and of course parameters) are re-estimated |

|Complete |The model is completely re-estimated |

Obviously, other options could be added in the future (for instance, identification of new outliers on the last period(s)).

When the refresh option has been selected, Demetra+ automatically goes to the suitable time series provider(s) to ask for the updated observations; the new estimations are done on those series (using the previous models, modified by the chosen option).

In next papers, we will describe in details and in a more formal way all the features of Demetra+.

-----------------------

[1] Closed panels can be re-opened through the main menu commands: Workspace->View->...

[2] Description from G. Caporello, A. Maravall “PROGRAM TSW, REVISED REFERENCE MANUAL,” July 2004. TSW program be downloaded form webpage .

[3] D. Findley, B. Monsell, W. Bell, M.Otto, B. Chen, New Capabilities and Methods of the X-12-ARIMA Seasonal Adjustemnt Program, pp.22.

[4] Definition taken from: X-12-ARIMA Reference Manual, p. 55,

[5] D. Findley, B. Monsell, W. Bell, M.Otto, B. Chen, New Capabilities and Methods of the X-12-ARIMA Seasonal Adjustemnt Program, pp.26.

[6] Findley, Monsell, Shulman and Pugh (1990), Sidings Spans Diagnostics for Seasonal and Related Adjustments, Bureau of the Census.

[7] It is not possible to drop a TramoSeats specification in an X12 output and vice versa.

[8] For the moment, those options are only available for multi-processing.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download