- University of Gothenburg
- School of Business, Economics and Law
- Department of Economics
- Units and Centra
- Statistical Research Unit
- Software
Software
Computer software development by the Statistical Research Unit at the University of Gothenburg, Gothenburg, Sweden. By downloading software you agree to the license terms described here
1. Outbreak DetectionThe programs for outbreak detection give an alarm signal when the process changes from a constant regression to a monotonically increasing one. This is the case when an outbreak of an epidemic disease starts. The version below is designed for the Poisson distribution. The Outbreak Detection program is freeware. It runs in Microsoft Excel for Windows. Click below to download. Updated 2011-02-22. Outbreak Detection P Instructions.pdf
For questions please contact Linus Schiöler.
2. Outbreak DetectionP MultivariateThe programs for outbreak detection monitors two processes and give an alarm signal when one of the processes changes from a constant regression to a monotonically increasing one. This is the case when an outbreak of an epidemic disease starts. The version below is designed for the Poisson distribution. The Outbreak Detection Multivariate program is freeware. It runs in Microsoft Excel for Windows. Click below to download. Updated 2011-02-22. Outbreak Detection P Multi Instructions.pdf
For questions please contact Linus Schiöler.
3. Unimodal Regression (Uregr)Uregr.sas is a program for SAS which computes maximum likelihood estimates of four different regressions: monotonically increasing and monotonically decreasing, u-shaped and inversely u-shaped. It produces the statistics necessary to test the hypothesis of the existence of a turn. The Uregr program is freeware. It runs within the Base SAS® software, which is available for many different computer platforms and operating systems. It has been tested on SAS 8.02, 9.1 and 9.2 on Microsoft Windows. The Uregr program can be downloaded by clicking the link below. For questions please contact Marianne Frisén (Marianne.Frisen@statistics.gu.se) References:
4. Turn DetectionChanges from an increasing regression to a decreasing is of interest to detect peaks. Troughs are detected by simple transformation. Likelihood based surveillance is used to give the alarm signals. References: Shiryaev-Roberts nonparametric (SRnp)SRnp.sas is a program for SAS. It contains further developments of Ureg.sas. It computes a non-parametric alarm statistic for detection of a turning point. SRnp.sas uses the Shiryaev-Roberts approach, which means that the alarm statistic is an unweighted sum of partial maximum likelihood ratios, L(1,s)+L(2,s)+...+L(s,s). The SRnp program is freeware. It runs within the Base SAS® software, which is available for many different computer platforms and operating systems. It has been tested on SAS 8.02, 9.1 and 9.2 on Microsoft Windows. A copy of the SRnp program can be dowloaded by clicking the link below. For questions please contact Marianne Frisén (Marianne.Frisen@statistics.gu.se)
5. Surveillance Evaluation (SE)
Statistical surveillance (monitoring, change point detection) is used when we have a continual observation of a time series, with the goal of detecting an important change in the underlying process as soon as possible after it has occurred and at the same time keep the rate of false alarms at an acceptable level. The SE program is used to visualize different measure of performance in surveillance. The measures (suggested by Frisén (1992)) are the cumulative false alarm probability, the probability of successful detection and the predictive value of an alarm at different time points. These measures will give information on the influence of time and the different risks of false judgements involved when repeated decisions are made. Several surveillance methods (Shewhart, CUSUM, SRnp and Shiryaev-Roberts methods) are compared, using the measures above. The SE program allows you to evaluate the effect of different parameter values for different methods. Three functions are shown in three graphs. These are the probability of false alarm, the probability of successful detection and the predictive value. Different settings of parameters results in different functions and the results can be compared in the same graphs. The SE program is freeware. It is compiled for running on a computer with a Microsoft Windows operating system (SE 1.41 is tested under XP, Vista and Windows 7). Click below to download. Surveillance Evaluation 1.41. For questions please contact Marianne Frisén (Marianne.Frisen@statistics.gu.se).
Turn Detection for ExcelThe program Turn Detection performs the same calculations as SRnp.sas and LRnp.sas. It was developed in Visual Basic and Microsoft Excel, with user friendliness and accessibility in mind. The program basically looks like an ordinary Excel spreadsheet, but has an additional menu and forms for choosing the parameters. The Turn Detection program is freeware. It runs in Microsoft Excel for Windows. To download a copy of the Turn Detection program click below. Turn Detection Manual.For questions please contact Marianne Frisén (Marianne.Frisen@statistics.gu.se). Likelihood Ratio nonparametric (LRnp)LRnp.sas is a program for SAS. It computes a non-parametric alarm statistic for detection of a turning point, as does SRnp.sas. However, in LRnp.sas the time of the turn is treated as a random variable with constant intensity, which implies a Geometric distribution. This means that the alarm statistic is a weighted sum of partial maximum likelihood ratios, w1*L(1,s)+w2*L(2,s)+...+ws*L(s,s), where the weights wj, j=1, 2, ¿, The LRnp program is freeware. It runs within the Base SAS® software, which is available for many different computer platforms and operating systems. It has been tested on SAS 8.02 on Microsoft Windows. A copy of the LRnp program can be downloaded by clicking the link below. For questions please contact Marianne Frisén (Marianne.Frisen@statistics.gu.se)
|