I have been working on a script in R that will predict a number.
# Load Forecast library
library(forecast)
# Load dataset
bwi <- read.csv(file="C:/Users/nsoria/Downloads/AMS Globales/TEC_BWI.csv", header=TRUE, sep=';', dec=",")
# Create time series starting in January 2015
ts_bwi <- ts(bwi$BWI, frequency = 12, start = c(2015,1))
# Pull out the seasonal, trend, and irregular components from the time series
model <- stl(ts_bwi, s.window = "periodic")
# Predict the next 5 months of SLA
pred <- forecast(model, h = 5)
# Plot the results
plot(pronostico)
This output gives this
Somehow, the forecasted line is not linked with the actual values.
Question: How can I make the line linked from the last known value to the first forecasted value?
Edit 01/01: Here is the link where the CSV is located to reproduce the case.
You need to add your real time series to the predicted one like in the code below
pred_mod<-pred
ts_real<-pred$x
pred_mod$x<-ts(c(ts_real,pred$mean),frequency=12,start=c(2015,1))
plot(pred_mod)
here the result
Related
I have a doubt related to the forecast () function from the package Forecast.
I am using this function for forecasting the closing price of a stock given an ARIMAX model (with xreg). The doubt is: when it is forecasting, the closing price at time t depends on the external regressors at time t-1 or it (closing price) depends on the external regressors at time t?
In other words, today I still don't know the high price (i.e.) so the closing price of today cannot depend on the high price of today, but on the one of yesterday.
This function works like that or in a different way?
I hope I have been clear. Thanks!
you can setup the function to work like this yes! Though there are some steps to take:
lag the regressor as you want yesterdays value to explain todays
clean values without regressor (first value of timeseries got no regressor as it will be used for the second value of the ts)
build the regressor for prediction
model and predict
Below I wrangled something together from a few links that shows how it can be done and thus should explain how prediction with regressor in your case works with forecast:
library(quantmod)
library(forecast)
library(dplyr)
# get some finance data to play with
quantmod::getSymbols("AAPL", from = '2017-01-01',
to = "2018-03-01",warnings = FALSE,
auto.assign = TRUE)
# I prefer working with df and then convert to ts objects later
new_AAPL <- as.data.frame(AAPL)%>%
# select close values and lag high values
dplyr::transmute(AAPL.Close,
AAPL.High = lag(AAPL.High)) %>%
# keep only complete values
dplyr::filter(across(everything(), ~!is.na(.x)))
# set up new time series, regressor (watch the starting points)
AAPL.Close <- ts(new_AAPL$AAPL.Close, start = as.Date("2017-01-04"), frequency = 365)
AAPL.High <- ts(new_AAPL$AAPL.High, start = as.Date("2017-01-04"), frequency = 365)
# set up the future regressor (last value of original high values
AAPL.futureg <- ts(as.data.frame(AAPL)$AAPL.High[291], start = as.Date("2018-03-02"), frequency = 365)
# I will use a arima model here
modArima <- forecast::auto.arima(AAPL.Close, xreg=AAPL.High)
# forecast with regressor
forecast::forecast(modArima, h = 1, xreg = AAPL.futureg)
Here is where I got the infos from:
https://www.codingfinance.com/post/2018-03-27-download-price/
https://stats.stackexchange.com/questions/41070/how-to-setup-xreg-argument-in-auto-arima-in-r
I have player over time data that is missing player counts over several years. I'm trying to fill in/predict the missing player count data over different intervals.
Data available here: https://1drv.ms/u/s!AvEZ_QPY7OZuhJAlKJN89rH185SUhA
I'm following the instructions below that use KalmanRun to impute the missing values. I've tried 3 different approaches to transforming the data- using an xts object, and 2 approaches to converting it into time series data
https://stats.stackexchange.com/questions/104565/how-to-use-auto-arima-to-impute-missing-values
require(forecast)
library(xts)
library(anytime)
library(DescTools)
df_temp = read.csv("r_share.csv")
df_temp[['DateTime']] <- as.Date(strptime(df_temp[['DateTime']], format='%Y-%m-%d %H:%M:%S'))
3 approaches to convert data; xts seems to work best by returning non-zero data that is interpretable.
#Convert df_temp to TimeSeries object
df_temp = xts(df_temp$Players, df_temp$DateTime)
#df_temp = as.ts(log(df_temp$Players), start = start(df_temp$DateTime), end = end(df_temp$DateTime), frequency = 365)
#df_temp = ts(df_temp$Players, start = c(2013, 02, 02), end = c(2016, 01, 31), frequency = 365)
Fitting and plotting:
fit <- auto.arima(df_temp, seasonal = TRUE)
id.na <- which(is.na(df_temp))
kr <- KalmanRun(index(df_temp), fit$model, update = FALSE)
#?KalmanRun$tol
for (i in id.na)
df_temp[i] <- fit$model$Z %*% kr$states[i,]
plot(df_temp)
The expected output is data that mimics the variability seen in the actual data and is different for each interval, whereas the actual output is relatively stationary and unchanging (both intervals have nearly the same prediction).
It needs to be with model arima()?.
Maybe you could try with another model, developed by Facebook named Prophet.
Here you can find the guide and github page.
If I understood you want something like this:
# Import library
library(prophet)
# Read data
df = read.csv("C:/Users/Downloads/r_share.csv",sep = ";")
# Transform to date
df["DateTime"] = as.Date(df$DateTime,format = "%d/%m/%Y")
# Change names for the model
colnames(df) = c("ds","y")
# call model
m = prophet(df)
# make "future" just one day greater than past
future = make_future_dataframe(m,periods = 1)
# predict the points
forecast = predict(m,future)
# plot results
plot(m,forecast)
I'm trying to create a forecast using an exponential smoothing method, but get the error "nonseasonal data". This is clearly not true - see code below.
Why am I getting this error? Should I use a different function (it should be able to perform simple, double, damped trend, seasonal, Winters method)?
library(forecast)
timelen<-48 # use 48 months
dates<-seq(from=as.Date("2008/1/1"), by="month", length.out=timelen)
# create seasonal data
time<-seq(1,timelen)
season<-sin(2*pi*time/12)
constant<-40
noise<-rnorm(timelen,mean=0,sd=0.1)
trend<-time*0.01
values<-constant+season+trend+noise
# create time series object
timeseries<-as.ts(x=values,start=min(dates),end=max(dates),frequency=1)
plot(timeseries)
# forecast MAM
ets<-ets(timeseries,model="MAM") # ANN works, why MAM not?
ets.forecast<-forecast(ets,h=24,level=0.9)
plot(ets.forecast)
Thanks&kind regards
You should use ts simply to create a time series from a numeric vector. See the help file for more details.
Your start and end values aren't correctly specified.
And setting the frequency at 1 is not a valid seasonality, it's the same as no seasonality at all.
Try:
timeseries <- ts(data=values, frequency=12)
ets <- ets(timeseries, model="MAM")
print(ets)
#### ETS(M,A,M)
#### Call:
#### ets(y = timeseries, model = "MAM")
#### ...
The question in your comments, why ANN works is because the third N means no seasonnality, so the model can be computed even with a non-seasonal timeseries.
I have fitted a TBATS model around my seasonal time-series data and used the forecast package to obtain predictions. My R code is:
library("forecast")
data = read.csv("data.csv")
season_info <- msts(data,seasonal.periods=c(24,168))
model <- tbats(season_info)
forecasted <- forecast.tbats(best_model,h=24,level=90)
forecasted
Now, I have a variable called 'forecasted' that outputs as such:
> forecasted
Point Forecast Lo 90 Hi 90
6.940476 5080.641 4734.760 5426.523
6.946429 5024.803 4550.111 5499.496
6.952381 4697.625 4156.516 5238.733
6.958333 4419.105 3832.765 5005.446
6.964286 4262.782 3643.528 4882.037
6.970238 4187.629 3543.062 4832.196
6.976190 4349.196 3684.444 5013.947
6.982143 4484.108 3802.574 5165.642
6.988095 4247.858 3551.955 4943.761
6.994048 3851.379 3142.831 4559.927
7.000000 3575.951 2855.962 4295.941
7.005952 3494.943 2764.438 4225.449
7.011905 3501.354 2760.968 4241.739
7.017857 3445.563 2695.781 4195.345
I need to gather the forecasted values from the column 'Forecast' and store it in a CSV file. I tried to read the page for the TBATS and 'forecast' method online, but they do not say how a particular column of forecasted values could be extracted, ignoring the other columns such as 'Hi' 'Lo' and 'Point'.
I'm looking for this output in my CSV:
hour,forecasted_value
0,5080.641
1,5024.803
2,4697.625
...
They are stored in three parts. You can look at the object structure with str(ret):
library(forecast)
fit <- tbats(USAccDeaths)
ret <- forecast(fit)
ret$upper # Upper interval
ret$lower # Lower interval
ret$mean # Point forecast
You can obtain the output shown by using print():
library("forecast")
data = read.csv("data.csv")
season_info <- msts(data,seasonal.periods=c(24,168))
model <- tbats(season_info)
forecasted <- forecast.tbats(best_model,h=24,level=90)
dfForec <- print(forecasted)
this will give you the data.frame, now you can pick out the columns you want, ie. dfForec[, 1] for only the point-forecast, then use write.csv(dfForec[, 1, drop = FALSE], ...) to write it to a flat file.
use mean function for getting your Point Forecast
library("forecast")
data = read.csv("data.csv")
season_info <- msts(data,seasonal.periods=c(24,168))
model <- tbats(season_info)
forecasted <- (forecast.tbats(best_model,h=24,level=90))$mean
or
forecasted$mean
I have data of the form SaleDateTime = '2015-01-02 23:00:00.000' SaleCount=4.
I'm trying to create an hourly forecast for the next 12 hours, using the code below.
I'm new to forecasting and could definitely appreciate some advice.
I'm trying to partition the data, train a model, plot the forecast with x axis of the form '2015-01-02 23:00:00.000', and test the accuracy of the model on a test time series.
I'm getting the error message below, when I try to run the accuracy as shown. Does anyone know why I'm getting the error message below?
When I run the plot as shown below it has an x axis from 0 to 400, does anyone know how to show that as something like '2015-01-02 23:00:00.000'? I would also like to narrow the plot to the last say 3 months.
My understanding is that if you don't specify a model for forecast, then it tries to fit the best model it can to the data for the forecast. Is that correct?
How do I filter for the same timeseries range with the forecast as the ts1Test that I'm trying to run accuracy on, is it something like ts(fcast2, start=2001, end = 8567) ?
Since I'm using the zoo package is the as.POSIXct step unnecessary, could I just do eventdata <- zoo(Value, order.by = SaleDateTime) instead?
library("forecast")
library("zoo")
SampleData<-SampleData
Value<-SampleData[,c("SaleDateTime","SaleCount")]
rDateTime<-as.POSIXct(SampleData$SaleDateTime, format="%Y-%m-%d %H:%M:%S")
eventdata <- zoo(Value, order.by = rDateTime)
##Partitioning data Training/Testing
ts1SampleTrain<-eventdata[1:2000,]
ts1Train<-ts(ts1SampleTrain$SaleCount, frequency=24)
ts1SampleTest<-eventdata[2001:28567,]
ts1Test<-ts(ts1SampleTest$SaleCount, frequency=24)
#Training Model
fcast2<-forecast(ts1Train,h=8567)
plot(fcast2)
accuracy(fcast2,ts1Test)
New Error:
Error in -.default(xx, ff[1:n]) : non-numeric argument to binary operator
To make your accuracy test run you should ensure that the length of your test data ts1Test and your forecasting horizon, h in fcast2<-forecast(ts1Train,h=8567) are of the same length. Now you have 26567 datapoints vs 8567.
Following your approach, the next toy example will work:
library(forecast)
library(zoo)
Value <- rnorm(1100)
rDateTime <- seq(as.POSIXct('2012-01-01 00:00:00'), along.with=Value, by='hour')
eventDate <- ts(zoo(Value, order.by=rDateTime), frequency = 24)
tsTrain <-eventDate[1:1000]
tsTest <- eventDate[1001:1100]
fcast<-forecast(tsTrain,h=100)
accuracy(fcast, tsTest)
ME RMSE MAE MPE MAPE MASE ACF1
Training set -2.821378e-04 9.932745e-01 7.990188e-01 1.003861e+02 1.007542e+02 7.230356e-01 4.638487e-02
Test set 0.02515008 1.02271839 0.86072703 99.79208174 100.14023919 NA NA
Concerning your other two questions:
Use of POSIX timestamps and zoo package. You don't need them to
use forecast. ts(Value, frequency) would suffice.
Plotting time series object with datetimes as your labels. The
following code snippet should get you started in this direction. Look for
axis function that provides the desired behavior:
par(mar=c(6,2,1,1)) # bottom, left, top, right margins
plot(tsTrain, type="l", xlab="", xaxt="n")
axis(side=1, at=seq(1,1000,100), label=format(rDateTime[seq(1,1000,100)], "%Y-%m-%d"), las=2)