18.Time series04.Autoregression - sporedata/researchdesigneR GitHub Wiki

1. Use cases: in which situations should I use this method?

A time series is a collection of data points measured multiple times over a given period - a sequence taken at repeated and successive equally spaced points in time. Meanwhile, autoregression is a time series model used to forecast a range of time series problems accurately. Autoregression uses observations from previous time steps to predict the value at the next time step.

The key concept of autoregression is that the observations are associated - if you have a value for today, you can predict the value for tomorrow, which will then predict the subsequent day, and so. In other words, an autoregression model enables us to predict the current value or variable of interest using a linear combination of past values of the variable - the model assumes that the past values of the time series are affecting its current value (autocorrelation). Now, you cannot use regular regression models like GLM since the observations are not independent -- the latter is an assumption of GLM.

  • Autoregression can be used to create counterfactuals.

2. Input: what kind of data does the method require?

3. Algorithm: how does the method work?

Model mechanics

Time series analysis is used for datasets consisting of a series of data points displayed over time.

Reporting guidelines

Data science packages

Suggested companion methods

Learning materials

  1. Books

  2. Articles

4. Output: how do I interpret this method's results?

Mock conclusions or most frequent format for conclusions reached at the end of a typical analysis.

Tables, plots, and their interpretation

5. SporeData-specific

Templates

Data science functions

References

[1] Epskamp S, Waldorp LJ, Mõttus R, Borsboom D. The Gaussian graphical model in cross-sectional and time-series data. Multivariate behavioral research. 2018 Jul 4;53(4):453-80. [2] Molina LL, Angón E, García A, Moralejo RH, Caballero-Villalobos J, Perea J. Time series analysis of bovine venereal diseases in La Pampa, Argentina. PloS one. 2018 Aug 6;13(8):e0201739. [3] Maleki M, Mahmoudi MR, Wraith D, Pho KH. Time series modelling to forecast the confirmed and recovered cases of COVID-19. Travel medicine and infectious disease. 2020 Sep 1;37:101742. [4] Ward MP, Iglesias RM, Brookes VJ. Autoregressive models applied to time-series data in veterinary science. Frontiers in veterinary science. 2020:604. [5] Chen D.Uncertain regression model with autoregressive time series errors . Soft Computing. 2021 Dec;25(23):14549-59. [6] Taira K, Hosokawa R, Itatani T, Fujita S. Predicting the Number of Suicides in Japan Using Internet Search Queries: Vector Autoregression Time Series Model. JMIR public health and surveillance. 2021 Dec 3;7(12):e34016. [7] Tobias S, Grant CJ, Laing R, Arredondo J, Lysyshyn M, Buxton J, Tupper KW, Wood E, Ti L. Time-series analysis of fentanyl concentration in the unregulated opioid drug supply in a Canadian setting. American Journal of Epidemiology. 2022 Feb;191(2):241-7.

⚠️ **GitHub.com Fallback** ⚠️