SPSS command: SPSS CREATE
Create time series: differences, lags and smoothers
CREATE builds a new series from an existing one with a time-series function: a difference to remove a trend, a lag or lead to line up a predictor, or a moving average or T4253H smoother to reveal the underlying movement.
Free · No account · Files stay on your device · Updated 18 October 2026
What it tells you
Crosstabs offers nine SPSS CREATE functions, previews the result and lets you download it as CSV without changing your file.
- What the series looks like after differencing, smoothing, lagging or accumulating, shown next to the original.
- How many values the function loses at the start or end of the series.
- A smoothed trend line (moving average, running median or T4253H) for reporting a noisy tracker.
When to use it
Use it when
- You need a lagged or differenced variable for a regression, AREG or FIT run.
- A monthly tracker is noisy and you want a 3-month moving average or a robust smooth for a chart.
- You want year-on-year changes: a seasonal difference of order 1 with periodicity 12.
Use something else when
- You want the new variable inside your open dataset: CREATE here produces a preview and CSV, not a new column.
- You want to fill missing values: use Replace missing values, or the time axis's linear interpolation option.
- You want a model-based trend or seasonal adjustment: use seasonal decomposition or exponential smoothing.
How to run it in Crosstabs
- 01
Open Create time series
In Statistics mode, choose Create time series (CREATE) in the Forecasting group.
- 02
Define the time axis
Group respondent rows into periods or use each row as a time point. A Seasonal difference uses the periodicity set here.
- 03
Choose the Series and Function
Pick the Series and a Function: Difference, Seasonal difference, Centered moving average, Prior moving average, Running median, Cumulative sum, Lag, Lead or Smoothing (T4253H).
- 04
Set the Order or Span
Differences, lags and leads take an Order; moving averages and running medians take a Span. Cumulative sum and T4253H need neither.
- 05
Run, preview and download
Click Run to see the preview table and chart, then use Download new series (CSV) to keep all values, named in SPSS style such as sales_ma3.
Worked example
Computed by the Crosstabs engine from the synthetic data below when this page was built — not typed in.
Data
48 synthetic monthly values (January 2022 to December 2025) of a brand-tracker metric in percent: a level of 40 rising 0.25 points a month, a fixed seasonal pattern that peaks in December and dips in July, and normal noise (SD 1.1) from a fixed seed, so every build produces the same series.
| Time point | awareness | awareness_ma3 | awareness_sdiff1 | awareness_t4253h |
|---|---|---|---|---|
| 2022-01 | 41.8 | — | — | 41.55 |
| 2022-02 | 39.7 | 40.60 | — | 40.78 |
| 2022-03 | 40.3 | 40.53 | — | 40.16 |
| 2022-12 | 47.1 | 45.27 | — | 44.66 |
| 2023-01 | 44.5 | 44.80 | 2.70 | 44.63 |
| 2023-02 | 42.8 | 43.87 | 3.10 | 44.40 |
Each created column is a separate run of the panel (centered moving average span 3, seasonal difference order 1, T4253H). — marks a missing value.
How to read it
The centered moving average loses one value at each end, so it has 46 values; the 12-month seasonal difference loses the first year and keeps 36. Its mean of 2.95 points is the average year-on-year change, about 12 × 0.25 = 3 by construction. T4253H fills every time point (48 values) with a robust smooth.
Output tables explained
- Created Series (preview)
- Time point, original value and new value. Series longer than 24 points show the first and last 12; the CSV has every value. Blank cells are where the function needs values outside the series.
- Original and created series
- A line chart of the original and new series on the same time axis.
Assumptions and notes
- Time points are equally spaced; a missing input inside a function's window makes that output missing, except Cumulative sum, which carries the total past it.
- A Seasonal difference needs a periodicity of 2 or more.
- Even spans for a centered moving average are centred the SPSS way, averaging two adjacent windows.
How the results are validated
Twelve function settings (cumulative sum, first and second differences, a 12-period seasonal difference, lag 2, lead 3, centred moving averages of span 5 and 4, a prior moving average of span 3, running medians of span 5 and 4, and T4253H) match numpy and pandas rolling computations to 1e-11; T4253H is checked against an independent Python implementation of the SPSS manual's description. Validation evidence
Questions
Is the new series added to my data?
No. Unlike SPSS CREATE, the panel does not add a column: it previews the series and offers it as a CSV you can merge back if needed.
What is T4253H?
A robust compound smoother: running medians of 4, 2, 5 and 3, then a Hanning moving average, applied again to the residuals and added back. It resists outliers better than a moving average.
Centered or prior moving average?
A centered average uses values on both sides and suits charts of the past. A prior average uses only earlier values, so it can serve as a simple forecast or predictor.
How do I get year-on-year change?
Set Periodicity to 12 (or 4 for quarters) in step 1, then choose Seasonal difference with order 1. Each value is this period minus the same period a year earlier.
Related procedures
- Seasonal DecompositionSEASON
- Autocorrelations (ACF and PACF)ACF
- Replace Missing ValuesRMV
- Forecast Goodness of FitFIT
- Autoregression (AREG)AREG
- Free SPSS alternative: full coverage list
Run Create Time Series on your own data
Open a .sav, CSV or Excel file. Everything runs in your browser; the file never leaves your device.
SPSS is a trademark of IBM Corp. Crosstabs is not affiliated with or endorsed by IBM.