arima

问题 I Have an R code that helps me to know at what seed when I use arima.sim() function to simulate ARIMA(1, 0, 0) it will actually simulate ARIMA of order 1, 0, 0 when auto.arima() function is employed for a check. MWE library(forecast) SEED_vector <- 1:10 arima_order_results <- data.frame() flag <- TRUE i <- 1 seed_out <- c() while(flag){ set.seed(SEED_vector[i]) ar1 <- arima.sim(n = 20, model=list(ar=0.8, order = c(1, 0, 0)), sd = 1) ar2 <- auto.arima(ar1, ic = "aicc") if(all(arimaorder(ar2)=

问题 I keep getting an error when fitting my ARIMA to the data, 'data' must be of a vector type, was 'NULL'. library(forecast) foo <- read.csv("https://nofile.io/g/0qrJl41nhf3bQQFjBmM6JurzGJFQSioCTGEzZhWVl9zA1kXnAJsCsSsxN1ZN7F4D/data.csv/") data <- data.frame(year, Car) data <- ts(data[,2],start = c(1990,1),frequency = 1) plot(data) plot(diff(data),ylab='Differenced Car Usage') plot(log10(data),ylab='Log (Car Usage)') plot(diff(log10(data)),ylab='Differenced Log (Tractor Sales)') par(mfrow = c(1,2

问题 Using an arima.sim() function to simulate time series data that follows a particular ARIMA model requires a lot of trials of this nature: library(forecast) set.seed(1) ar1 <- arima.sim(n = 10, model=list(ar=0.2, order = c(1, 0, 0)), sd = 1) ar2 <- auto.arima(ar1, ic ="aicc") ar2 One needs to be changing the seed integer until the desired result is archived. I now think of instead of changing the seed integer manually and checking with auto.arima() function I should automate the seeds with a

问题 How would I calculated standartized residuals from arima model sarimax function? lets say we have some basic model: import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns sns.set(style='ticks', context='poster') from statsmodels.tsa.statespace.sarimax import SARIMAX from statsmodels.tsa.seasonal import seasonal_decompose import seaborn as sns #plt.style.use("ggplot") import pandas_datareader.data as web import pandas as pd import statsmodels.api as sm

问题 Often times I make use of arima.sim() function to simulate ARIMA model and later discover through auto.arima() function that the ARIMA model simulated is not the same with what I specified in arima.sim() function. I went further to investigate to know how does the arima.sim() fair in simulating ARIMA model by simulating same ARIMA model with the same arima.sim() detail good number of times and then check each out with auto.arima() here. result <- matrix(NA_integer_, nrow = 10, ncol = 3)

问题 I have been trying to use aTSA and Forecast package together and noticed that the Arima() function works but the forecast() give error. Does anyone have a solution for this or encountered this? I am especially trying to use stationary.test() from aTSA and that was the main reason I called the library. error: Error in forecast(.) : 'object' should be 'Arima' or 'estimate' class estimated from arima() or estimate() As soon as I removed aTSA, the above worked. fitArima_CO <- Arima(train_CO,

问题 I'm trying to get a "fuzzy" prediction of a timeseries, using an SARIMA model My training set is prices_train , and the model is built as follows: model_order = (0, 1, 1) model_seasonal_order = (2, 1, 1, 24) model = sm.tsa.statespace.SARIMAX( prices_train, order=model_order, seasonal_order=model_seasonal_order) model_fit = model.fit(disp=0) I know I can get a point forecast using this instruction: pred = model_fit.forecast(3) But I don't want a point forecast, I want a confidence interval of

问题 I try to get ARIMA configuration some faster that I acctually do. So I use a Iterate method to compare all ARIMA combinations to select better. For that I create a function to Iterate: def difference(dataset, interval=1): diff = list() for i in range(interval, len(dataset)): value = dataset[i] - dataset[i - interval] diff.append(value) return np.array(diff) # invert differenced value def inverse_difference(history, yhat, interval=1): return yhat + history[-interval] # evaluate an ARIMA model

问题 I want to estimate parameters for an ARIMA model. I do this in python with the arima function. Now, I want to remove the non significant lags. For instance, I only want the lags 1 and 3. But by order I can only give the total lags. (Hence, if I say p=3, then I get lag 1, 2 and 3) How can I solve this? model = ARIMA(R_bel, order=(3,0,1)) model_fit = model.fit(disp=0) print(model_fit.summary()) Thank you 回答1: If you want only specific list of lags like 1 & 3 as AR components, then you can do

问题 I'm trying to build old school model using only auto regression algorithm. I found out that there's an implementation of it in statsmodel package. I've read the documentation, and as I understand it should work as ARIMA. So, here's my code: import statsmodels.api as sm model = sm.tsa.AutoReg(df_train.beer, 12).fit() And when I want to predict new values, I'm trying to follow the documentation: y_pred = model.predict(start=df_test.index.min(), end=df_test.index.max()) # or y_pred = model