Convert txt file to time series and conduct trend and wavelet analyses
I have many .txt files all with the same format, one column for date (yyyy/mm/dd), another column for discharge (numerical values from 0.00 to 500.00), and two coulmns that I do not need. I am working in R.
I am looking to import this data with values for date and discharge, possibly convert the data to a time series, and then run Mann-Kendall trend tests and linear regressions on the discharge values.
I am interested in testing for trend on the full period of record in my data which is longer than 50 years, but I would like to test for trend by day of year. For example, I would be looking at discharge on January 1st for every year from 1950 to 2012. The I would want to test Jan 2 discharge for the period 1950 to 2012. How would I parse the data into data for just the day of the year for 52 years in order to run a trend test for each day of the year?
If I were also interested in only looking at trend in day of year discharge for example 1960 to 2000, how could I go about doing this?
Thank you very much, your help will be very greatly appreciated!