Skip to content

Latest commit

 

History

History
102 lines (61 loc) · 2.45 KB

README.md

File metadata and controls

102 lines (61 loc) · 2.45 KB
Raw

MwStatistics

MwStatistics allows you to easily calculate seven common statistics on time series data using a moving window.

Getting Started

May submit to CRAN. For now, refer to "Installing" to access the function.

Usage

General Usage.

MwStatistics(dat = NULL, TimeVarCol = 1, ColA = 2, ColB = 3, win = NULL,
stat = NULL, overlap = 0, genplot = T, genDataFrame = F)

Example

MwStatistics(dat = data.frame(replicate(10,sample(0:10,1000,rep=TRUE))),
TimeVarCol = 1, ColA = 2, ColB = 3, win = 2,
stat = "Correlation", genplot = T, genDataFrame = F, overlap = 0.5)

Availible statistics:

  • "Correlation" = Correlation

  • "P-Value" = P-Value

  • "SDevX" = Standard Deviation of variable A

  • "SDevY" = Standard Deviation of variable B

  • "SErrorSlope" = Standard Error of the slope

  • "SErrorIntercept" = Standard Error of the intercept

  • "Intercept" = Intercept (Linear regression)

  • "Slope" = Slope (Linear regression)

Arguments:

  • dat

    • Dataset name

  • TimeVarCol

    • Column number of time series variable

  • ColA

    • Column number of first variable

  • ColB

    • Column number of second variable

  • win

    • Window size

  • stat

    • Statistic to calculate

  • overlap

    • How much the windows overlap

  • genplot

    • Generate a plot of recorded statistic

  • genDataFrame

    • Generate dataframe with recorded statistic values

Installing

Until I submit to CRAN (if I choose to), the most reliable way to use this package is to simply download it from GitHub and install locally. Unfortunately, I have not quite figured out how to install directly into R via GitHub, so if you are willing to help out in the reguard, please let me know! My email is [email protected]

You can check that it installed correctly using:

?MWStat::MwStatistics

Running The Function

Fill in the correct information, replacing the sample information provided and run by selecting and clicking command enter.

Contributing

If you find any improvements, please email me at [email protected].

Authors

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE file for details

Acknowledgments

Inspiration from MwCor from Astrochron.