search for: stationary

This is not exactly an R question but the R code below may make my question more understandable. If one plots sin(x) where x runs from -pi to pi , then the curve hovers around zero obviously. so , in a"stationary in the mean" sense, the series is stationary. But, clearly if one plots the acf, the autocorrelations at lower lags are quite high and, in the "box jenkins" sense, this series is clearly not stationary in terms of its acf. so, i'm confused in terms of what ithe statistical def...

Hi all, Can anyone clear my doubts about what conclusions to take with the following what puts of some time series tests: > adf.test(melbmax) Augmented Dickey-Fuller Test data: melbmax Dickey-Fuller = -5.4075, Lag order = 15, p-value = 0.01 alternative hypothesis: stationary Warning message: p-value smaller than printed p-value in: adf.test(melbmax) >adf.test(melbmax,k=0,alternative="stationary") Augmented Dickey-Fuller Test data: melbmax Dickey-Fuller = -24.4069, Lag order = 0, p-value = 0.01 alternative hypothesis: stationary Warning...

...ing a, I use the following codes result1 = arima(ts(myinput),order = c(p,d,q),seasonal = list(order=c(P,D,Q))) result2 = forecast(result1,12) plot(result2) Now, by using R code... 1) How can I determine if my data is statitonary or not ? (trend & seasonal effects) 2) If not, how can I make it stationary ? 3) Is arima() function used only on STATIONARY data ? Or does it first determine if the data is stationary or not and makes it stationary ? (if it is non-stationary) 4) I tried different parameter values in arima() function, but every parameter gave very different results :(( . I even found &...

Short question: Is it possible to use statistical tests, like the Augmented Dickey-Fuller test, in functions with for-loops? If not, are there any alternative ways to scale measures? Detailed explanation: I am working with time-series, and I want to flag curves that are not stationary and which display pulses, trends, or level shifts. >df DATE ID VALUE2012-03-06 1 5.672012-03-07 1 3.452012-03-08 1 4.562012-03-09 1 20.302012-03-10 1 5.102012-03-06 2 5.672012-03-07 2 3.452012-03-08 2 4.562012-03-09 2 5.282012-03-10 2 5.1020...

...5.23 5.23 5.24 5.24 5.24 5.25 5.24 5.24 5.25 5.26 5.26 5.26 5.26 5.26 5.26 5.26 5.27 [145] 5.27 5.26 5.27 5.27 5.28 5.29 5.29 5.29 5.29 5.30 5.30 5.30 5.31 5.31 5.31 5.32 5.32 5.33 [163] 5.33 Now I want to conduct a test for stationarity using ADF test : > adf.test((data[,1]), "stationary", 0) Augmented Dickey-Fuller Test data: (data[, 1]) Dickey-Fuller = -3.7351, Lag order = 0, p-value = 0.02394 alternative hypothesis: stationary But surprisingly it leads towards rejestion of NULL [p-value is less than 0.05], i.e. indicates a possible stationary series. Howe...

...cal values: 0.10 0.05 0.025 0.01 0.347 0.463 0.574 0.739 ---- Statistic for the null hypothesis of trend stationarity: 0.13 Critical values: 0.10 0.05 0.025 0.01 0.119 0.146 0.176 0.216 ---- Lag truncation parameter: 1 CONCLUSION: Reject Ho at 0.05 sig level - Level Stationary Fail to reject Ho at 0.05 sig level - Trend Stationary > kpss.test(df,null = c("Trend")) KPSS Test for Trend Stationarity data: tsdata[, 6] KPSS Trend = 0.1298, Truncation lag parameter = 1, p-value = 0.07999 CONCLUSION: Fail to reject Ho - Trend Stationary...

...to make a covariogram of spatial data ? 2) does anyone know if there exist the possibility in R of performing kriging with an arbitrary covariance function ? If I have well understood the Krig procedure offers only three possibilities (expcov, gauscov and sphercov) that are the common models for stationary and isotropic processes. Thanks, Paola. -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not t...

...nerate random walk models in R? Basically, I need an AR(1) model with the phi^1 value equal to 1: Yt = c + Yt-1 + E where E is random white noise. I tried using the arima.sim command: arima.sim(list(ar=c(1)), n = 1000, rand.gen = rnorm) but got this error since the model I am generating is not stationary: Error in arima.sim(list(ar = c(1)), n = 1000, rand.gen = rnorm) : 'ar' part of model is not stationary I found arima.sim sufficient for generating stationary models, but how about non-stationary models? Thanks again for your help.

Hi, I am trying to test whether a series is return series stationary, but before proceeding I wanted to make sure I understand correctly how to use the adf.test function and interpret its output... Could you please let me know whether I am correct in my interpretations? ex: I take x such as I know it doesn't have a unit root, and is therefore stationary 1/ &gt...

Hi all, I was running the adf test in R. CODE 1: adf.test(data$LOSS) Augmented Dickey-Fuller Test data: data$LOSS Dickey-Fuller = -1.9864, Lag order = 2, p-value = 0.5775 alternative hypothesis: stationary CODE 2: adf.test(diff(diff(data$LOSS))) Augmented Dickey-Fuller Test data: diff(diff(data$LOSS)) Dickey-Fuller = -6.9287, Lag order = 2, p-value = 0.01 alternative hypothesis: stationary Is my interpretation correct: The original data( in code 1) is not stationary the twice diffe...

Hello! One note/question hier about specification of control-parameters in the lme(...,control=list(...)) function call: i tried to specify tne number of iteration needed via lme(....,control=list(maxIter=..., niterEM=...,msVerbose=TRUE)) but every time i change the defualt values maxIter (e.g. maxIter=1, niterEM=0) on ones specified by me, the call returns all the iterations needed until

Hi All, I am a newcomer to R. Could anyone explain me how to define link functions for either mu/sigma to allow for quadratic trends in the same, when fitting non-stationary GEV distributions? Thanks -- View this message in context: http://www.nabble.com/QUADRATIC-TREND-FOR-LINK-FUNCTIONS-ON-NON-STATIONARY-GEV-tp23360751p23360751.html Sent from the R help mailing list archive at Nabble.com.

Hi, I have a sample of size 100 from a function in interval [0,1] which can be assumed to come from a scalar-valued stationary Gaussian process. There are about 500 observation points in the interval. I need an effective and fast way to simulate from the Gaussian process conditioned on the available data. I can of course estimate the mean and 500x500 covariance matrix from data. I have searched both the Rsite and Internet...

I'm trying to understand how to use kpss.test() properly. If I have a level stationary series like rnorm() in the help page, shouldn't I get a small p-value with the null hypothesis set to "Trend"? The (condensed) output from kpss.test() for the two possible null hypotheses is given below. I don't see any significant difference between these results. > x &lt...

Hi I would like to use arima () to find the best arima model for y time series. The default in arima apparently is to use conditional sum of squares to find the starting values and then ML (as described on the help page). Now using the default may lead to error messages saying: "non-stationary ar part in CSS". When changeing the default to "ML" only the minimization works. As far as I understand, arima doesn't require stationarity, but apparently CSS does. Can anyone tell me what exactly the css method does? And why is CSS-ML the default in R? Out of efficiency rea...

Hi All, I am a newcomer to S/R. Could you please let me know how to model quadratic trends for the mul/sigl link functions when fitting non-stationary GEV distributions using the ismev package? Thanks Best Regards, Mohammad Ashrafuz Zaman PhD Candidate School of Engineering Building XC, Room 1.02 (Kingswood Campus) University of Western Sydney Locked Bag 1797, Penrith South DC NSW 1797 Australia Tel: 61-2-4736 0398 Mobile: 61-4-30977440 Email...

Dear R experts,   I'm running the following stationary bootstrap programming to find the parameters estimate of a linear model:     X<-runif(10,0,10) Y<-2+3*X a<-data.frame(X,Y) coef<-function(fit){   fit <- lm(Y~X,data=a)    return(coef(fit)) }  result<- tsboot(a,statistic=coef(fit),R = 10,n.sim = NROW(a),sim = "geom",orig....

...n I ensure that the function root(var) will run and display the output from all tests, and not just the last one? Thank you, b. root <- function(var) { #---Phillips-Perron PP.test(var, lshort = TRUE) PP.test(var, lshort = FALSE) #---Augmented Dickey-Fuller adf.test(var, alternative = "stationary", k = trunc((length(var)-1)^(1/3))) #---KPSS kpss.test(var, null = "Level", lshort = TRUE) kpss.test(var, null = "Trend", lshort = FALSE) }

...ally far from the behavior of my time series... in fact, there is a considerable gar between the last value of the series and the first forecast. My guess is that I'm doing something wrong. Here is what I do: >mods<-auto.arima(x[[1]],start.p=0,start.q=0,start.P=0,start.Q=0,stepwise=TRUE,stationary=FALSE) >ARIMA(2,1,4)(2,0,1)[12] with drift # the output Call: auto.arima(x = x[[k]], start.p = 0, start.q = 0, start.P = 0, start.Q = 0, stationary = FALSE, stepwise = TRUE) Coefficients: ar1 ar2 ma1 ma2 ma3 ma4 sar1 0.0639 -0.7820 -1.210...

...n many papers regarding time series analysis of acquired data, the authors analyze 'marginal distribution' (i.e. marginal with respect to time) of their data by for example checking 'cdf heavy tail' hypothesis. For i.i.d data this is ok, but what if samples are correlated, nonstationary etc.? Are there limit theorems which for example allow us to claim that for weak dependent, stationary and ergodic time series such a 'marginal distribution w.r. to time' converges to marginal distribution of random variable x_t , defined on basis of joint distribution for (x_1,&...