Arima

c<-auto.arima(fil[[1]],start.p=0,start.q=0,start.P=0,start.Q=0,stepwise=TRUE,stationary=FALSE,trace=TRUE)

system.time(fit.500<-auto.arima(ts, d = NA, D = NA, max.p = 2, max.q = 2,max.P = 1, max.Q = 1, max.order = 5,

start.p=0, start.q=0, start.P=0, start.Q=0,stationary = FALSE, ic = c("aic","aicc", "bic"),stepwise=FALSE,trace=TRUE))

tsdiag(b)

x<-scan("trace")

x_arima<-arima(x, order=c(1,0,1),seasonal=list( order=c(1,1,1),period=12));

b_arima<-forecast(x_arima)

#attributes(b)

#summary(a)0

ets(data)#exponential smoothening

accuracy(b_arima$fitted,b_arima$x)

predict continuously

#library

library(forecast)

require(graphics)

require(stats)

zro <- array(0, dim=c(1,400))

#constants

    Data_rate <- 20756 #byte per 5 mili sec

#Inputs

    gap<-50

    start<-0

    end<-200

    sum<-0

#read file 

input_file<-"riverbed.txt"

output_file<-"Predected_riverbed.txt"

unlink(output_file)

    trace=scan(input_file)

    size<-length(trace)

    arr<-trace[start:end]

    write(c(arr), file = output_file,append = TRUE, sep = "\n")

   while(end+gap<size)

        {

        #trace=scan("a")

        end2<-0

        arr<-trace[start:end]

        end2<-end+gap

        end3<-end+1

        arr2<-trace[end3:end2]

#forecast

        x_arima<-arima(arr, order=c(1,0,1),seasonal=list( order=c(1,1,1),period=16),method = c("CSS"))

        b_arima<-forecast(x_arima,gap)

        #accuracy(b_arima$fitted,b_arima$x)

        #attributes(b_arima)

        #summary(s_arima)

#framing

wframe <- array(0, dim=c(1,400))

framing <-c(b_arima$mean)

#framing <-arr2

i=1;

j=1;

count=0

    while (j <=gap){

            a<-framing[j]

             sum1=a

            if(sum1 < Data_rate){ wframe[i]=sum1;sum1=0;wframe[i+1:i+8]=0; i=i+8}            

            else if(sum1 == Data_rate){ wframe[i]=Data_rate;wframe[i+1:i+8]=0; sum1=0;i=i+8}    

            else if(sum1 > Data_rate){ 

                while(sum1>Data_rate) 

                {

                sum1=sum1-Data_rate

                wframe[i]=Data_rate

                i=i+1

                count=count+1

                }

                wframe[i]=sum1; sum1=arr[start+i];i=i+8-count-1; sum1=0;count=0}

        j=j+1;

#print(wframe[i])

#write(i, file = "ufhfui",append = TRUE, sep = "\n")

        }

#1/0

abc<-wframe

wframe <- array(0, dim=c(1,400))

for (k in 1:400) {

                dd=abc[k]

                #write(dd, file = "ufhfui",append = TRUE, sep = "\n")

                if(dd==0){zro[k]=5; wframe[k]<-0}

                else if (dd>0) {wframe[k]<-10; zro[k]=0}

                else if (dd<0) {wframe[k]<-10; zro[k]=0}

                }

#plot

        op <- par(mfrow = c(2, 1)) 

        plot(c(b_arima$mean),type ="o",col = "red",ylab="Frame Size",xlab="Frame Index",main="Video Frame Predection")

        lines(arr2,type ="o",col = "blue")

        legend("topright",c("Original","Predected"), cex=0.8,col=c("blue","red"), pch=21:22, lty=1:2);

       plot(c(wframe),type ="h",col = "red",ylab="Frame Size",xlab="Wireless Frame Index",main="Wireless Frames ")

       lines(c(zro),type ="h",col = "green")

        legend("topright",c("Occupied Frames","Empty Frames"), cex=0.8,col=c("red","green"), pch=21:22, lty=1:2);

        start<-start+gap

        end<-end+gap

        cat("Predected: ",end,"\n",file = "")

        #Sys.sleep(1)

        par(op)

    write(c(b_arima$mean), file = output_file,append = TRUE, sep = "\n")

        }

        cat("!@#$%^&*()_+!@#$%^&*()_+","\n","done","\n",file = "")

cat(" Bandwidth 10 MHz","\n",file = "")

cat(" Downlink  29","\n",file = "")

cat(" Uplink 18","\n",file = "")

cat(" Preamble 1","\n",file = "")

cat(" fft Size 1024","\n",file = "")

cat(" Modulation QPSK","\n",file = "")

cat(" Code Rate 0.75","\n",file = "")

cat(" Data Sub Carrier 720","\n",file = "")

cat(" No of Subchannels 30","\n",file = "")

cat(" Symbol Perslot 2","\n",file = "")

cat(" Symbol Period 102.9 micro sec","\n",file = "")

cat(" frame_duration 5 micro sec","\n",file = "")

cat(" headers 51 Slots","\n",file = "")

cat(" Loading Factor 0.00625(160 fraction of BW is available) ","\n",file = "")

cat(" Video Frame Rate 25 frame/sec","\n",file = "")

cat(" Symbol/frame 0.048591","\n",file = "")

cat(" No of slots 24","\n",file = "")

cat(" DL slot / frame 420","\n",file = "")

cat(" Useful slots 369","\n",file = "")

cat(" Bits/slot 72","\n",file = "")

cat(" Data rate per DL 26568","\n",file = "")

cat(" Data rate per 5 mili sec 20756.25 byte","\n",file = "")

library(forecast)

require(graphics)

require(stats)

    gap<-50

    start<-0

    end<-241

    trace=scan("file.txt")

    size<-length(trace)

    arr<-trace[start:end]

    unlink("data")

    write(c(arr), file = "data",append = TRUE, sep = "\n")

   while(end+gap<size)

        {

       #trace=scan("a")

        arr<-trace[start:end]

        end2=end+gap

        arr2<-trace[end+1:end2]

        #png(filename="start.png", height=600, width=600,bg="white")

        x_arima<-arima(arr, order=c(1,0,1),seasonal=list( order=c(1,1,1),period=16),method = c("CSS"))

        b_arima<-forecast(x_arima,gap)

        #accuracy(b_arima$fitted,b_arima$x)

        #attributes(b_arima)

        #summary(s_arima)

        #plot(forecast(x_arima,50),col = "red",main="Frame Trace Predection",ylab="Frame Size",xlab="Frame Index")

        plot(c(b_arima$mean),type ="b",col = "red",ylab="Frame Size",xlab="Frame Index")

        legend(length(x)*15,0.95,c("Original","Predected"), cex=0.8,col=c("red","blue"), pch=21:22, lty=1:2);

        lines(arr2,type ="b",col = "blue")

        #plot(arr2,type ="b",col = "blue")

        #lines(x,type ="b",col = "red")

        start=start+gap

        end=end+gap

        cat("Predected: ",start,"\n",file = "")

        write(c(b_arima$mean),"\t",arr2,file = "data",append = TRUE, sep = "\n")

        #dev.off()

        Sys.sleep(1)

        }

library(forecast)

require(graphics)

require(stats)

    start<-0

    end<-200

    trace=scan("trace200")

    size<-length(trace)

    while(end+50<size)

        {

        trace=scan("trace200")

        arr<-trace[start:end]

        end2=end+50

        arr2<-trace[start:end2]

        x_arima<-arima(arr, order=c(1,0,1),seasonal=list( order=c(1,1,1),period=16));

        b_arima<-forecast(x_arima,50)

        #accuracy(b_arima$fitted,b_arima$x)

        #attributes(b_arima)

        #summary(s_arima)

        #plot(forecast(x_arima,50),col = "red",main="Frame Trace Predection",ylab="Frame Size",xlab="Frame Index")

        plot(arr,type ="l",col = "red")

        lines(arr2,type ="o")

        start=start+50

        end=end+50

        cat(start,"___________________________________________________________________________","\n",file = "")

        #Sys.sleep(1)

        }

arima.sim(model, n=, rand.gen = rnorm, innov = rand.gen(n, ...),n.start = NA, start.innov = rand.gen(n.start, ...),...)

arima.sim(c(1,0,1), n=, rand.gen = rnorm, innov = rand.gen(n, ...),n.start = NA, start.innov = rand.gen(n.start, ...),...)

arima.sim(n = 63, list(ar = c(0.8897, -0.4858), ma = c(-0.2279, 0.2488)),sd = sqrt(0.1796))

garsim(n, phi, X = matrix(0, nrow = n), beta = as.matrix(0), sd = 1, family = "gaussian",transform.Xbeta = "identity", link = "identity", minimum = 0, zero.correction = "zq1", c = 1, theta = 0)

a=arrep(notation = "arima", phi = c(0.88780450), d = 0, theta = c(-0.64850050), Phi = c(0.05159824), D = 1,Theta = c(-0.67678839), frequency = 12)

a=arrep(notation = "arima", phi = c(1), d = 0, theta = c(1), Phi = c(1), D = 1,Theta = c(1), frequency = 12)

phi:p vector of autoregressive coefficient.            (0.8878,0.0027)

d:difference operator, implemented: d element of (0,1,2).            

theta:q vector of moving average coefficients.

Phi:P vector of seasonal autoregressive coefficients.

D:Seasonal difference operator, implemented: D element of(0,1,2).

Theta:Q vector of seasonal moving average coefficients.

frequency:The frequency of the seasonality (e.g. frequency = 12 for monthly series with annual periodicity).

auto.arima(x, d = NA, D = NA, max.p = 5, max.q = 5,max.P = 2, max.Q = 2, max.order = 5,start.p=2, start.q=2, start.P=1, start.Q=1,stationary = FALSE, ic = c("aic","aicc", "bic"),stepwise=TRUE, trace=FALSE,approximation=length(x)>100 | frequency(x)>12, xreg=NULL,test=c("kpss","adf","pp"), allowdrift=TRUE)

HoltWinters(x)    Computes Holt-Winters Filtering of a given time series

rwf(x)        Returns forecasts and prediction intervals for a random walk with drift model applied to x

ets(x)        Returns Exponential smoothing state space model for series x

accuracy(x)    Accuracy measures for forecast model (ME,RMSE,MAE,MPE,MAPE,MASE)

forecast    generic function for forecasting from time series

forecast.Arima    forecasts based on the results produced by ‘arima’

forecast.ts    forecasts using exponential smoothing state space models

auto.arima    Returns best ARIMA model according to either AIC, AICc or BIC value

arima.sim    Simulate from an ARIMA model

garsim        Simulate a time series using a general autoregressive model

arrep        Invert (invertible) SARIMA(p, d, q, P, D, Q) models to AR representation

Arima

arima

ar

meanf

splinef

thetaf

croston

ses

holt

hw

forecast(object, h=ifelse(object$arma[5]>1,2*object$arma[5],10),level=c(80,95), fan=FALSE, xreg=NULL,...)

forecast(object, h=10, level=c(80,95), fan=FALSE, ...)

forecast(object, h=10, level=c(80,95), fan=FALSE, ...)

data <- c(67, 81, 93, 65, 18, 44, 31, 103, 64, 19, 27, 57, 63, 25, 22, 150,31, 58, 93, 6, 86, 43, 17, 9, 78, 23, 75, 28, 37, 23, 108, 14, 137,69, 58, 81, 62, 25, 54, 57, 65, 72, 17, 22, 170, 95, 38, 33, 34, 68,38, 117, 28, 17, 19, 25, 24, 15, 103, 31, 33, 77, 38, 8, 48, 32, 48,26, 63, 16, 70, 87, 31, 36, 31, 38, 91, 117, 16, 40, 7, 26, 15, 89,67, 7, 39, 33, 58)

rmse=rmse <- function(obs, pred) sqrt(mean((b_arima$fitted-b_arima$x)^2))