forecasting

问题 With the forecast package, I have a time series that I would like ?auto.arima to automatically pick the orders but I would like to coerce seasonality. The defaults for the function allow for the seasonal argument to be set to TRUE , but that only allows the option for seasonality not a coercion. auto.arima(x, d=NA, D=NA, max.p=5, max.q=5, max.P=2, max.Q=2, max.order=5, max.d=2, max.D=1, start.p=2, start.q=2, start.P=1, start.Q=1, stationary=FALSE, seasonal=TRUE, ic=c("aicc", "aic", "bic"),

问题 A maths problem really I think... I have some historical data for some spreadsheet outputs along with the number of rows and columns. What I'd like to do is use this data to predict the peak memory usage and time taken based on the - known - row and columns. So, if no historical data exists then there will be no predictions. 1 or 2 historical values will be very inaccurate but I hope that given a wide enough variety of historical values, then a reasonably-accurate prediction could be made? I

问题 Is there a way to specify minimum or maximum possible values in a forecast done with ETS/ARIMA models? Such as when forecasting a trend in % that can only go between 0% and 100%. I am using R package forecast (and function forecast ). 回答1: If your time series y has a natural bound [a, b] , you should take a "logit-alike" transform first: f <- function (x, a, b) log((x - a) / (b - x)) yy <- f(y, a, b) Then the resulting yy is unbounded on (-Inf, Inf) , suitable for Gaussian error assumption.

问题 In the following example, I am trying to use Holt-Winters smoothing on daily data, but I run into a couple of issues: # generate some dummy daily data mData = cbind(seq.Date(from = as.Date('2011-12-01'), to = as.Date('2013-11-30'), by = 'day'), rnorm(731)) # convert to a zoo object zooData = as.zoo(mData[, 2, drop = FALSE], order.by = as.Date(mData[, 1, drop = FALSE], format = '%Y-%m-%d'), frequency = 7) # attempt Holt-Winters smoothing hw(x = zooData, h = 10, seasonal = 'additive', damped =

问题 There is a relevant question about how ToUniversalTime works. But the relevant information about time conversions tells me there are relevant databases for time conversion in the past. My question is how can ToUniversalTime forecast and knows when will be the DST change in 2070 for example? 回答1: My question is how can ToUniversalTime forecast and knows when will be the DST change in 2070 for example? It can't. It doesn't have any special insight into the future of what the governments of our

问题 I'm having a similar problem to the questioners here had with the linear model predict function, but I am trying to use the "time series linear model" function from Rob Hyndman's forecasting package. Predict.lm in R fails to recognize newdata predict.lm with newdata totalConv <- ts(varData[,43]) metaSearch <- ts(varData[,45]) PPCBrand <- ts(varData[,38]) PPCGeneric <- ts(varData[,34]) PPCLocation <- ts(varData[,35]) brandDisplay <- ts(varData[,29]) standardDisplay <- ts(varData[,3]) TV <- ts

问题 I'm having a similar problem to the questioners here had with the linear model predict function, but I am trying to use the "time series linear model" function from Rob Hyndman's forecasting package. Predict.lm in R fails to recognize newdata predict.lm with newdata totalConv <- ts(varData[,43]) metaSearch <- ts(varData[,45]) PPCBrand <- ts(varData[,38]) PPCGeneric <- ts(varData[,34]) PPCLocation <- ts(varData[,35]) brandDisplay <- ts(varData[,29]) standardDisplay <- ts(varData[,3]) TV <- ts

问题 I have estimated a BEKK(1,1) model and now I want to have forecast values of the model. Following are the R codes to estimate the BEKK(1,1) model. > install.packages('MTS') > install.packages('rmgarch') > simulated <- simulateBEKK(2, 1000, c(1, 1)) ##prepare the matrix: > simulated <- do.call(cbind, simulated$eps) ##Estimate with default arguments: > estimated <- BEKK(simulated) > diagnoseBEKK(estimated) 回答1: Since your question (is it a question?) doesn't really make clear what you are

问题 I'm trying to make a Forecast function in Javascript based on the code from Excel, explained at https://support.office.com/en-US/article/FORECAST-function-50CA49C9-7B40-4892-94E4-7AD38BBEDA99 But I don't understand what is the x with a trait on top (also y) from the formula and so I don't know how to translate it in Javascript. Can someone could help me please? Thank you. 回答1: x with a trait on top is the mean of x (i.e. the average of all xs). the same is with y. if x values are 20,28,31,38

问题 I am working with air miles data set and i conducted three different tests to check for stationary in the time series data set Test 1: Using acf and pacf acf(airmiles) pacf(airmiles) After differentiating its seems most of the values lies in significance level now acf(diff(airmiles)) pacf(diff(airmiles)) Test 2: Using adf.test adf.test(airmiles,k=0,alternative = "stationary") Augmented Dickey-Fuller Test data: airmiles Dickey-Fuller = -1.1415, Lag order = 0, p-value = 0.8994 alternative