Outlier Detection Tasking Performance Category 4

library(tsoutliers)
july7cat4 <- read.csv("C:/Users/Rinor/Downloads/sgt/july7cat4.csv")
attach(july7cat4)
task <- Average.percentageTracked
timeSeries <- ts(task, frequency=1, start=c(0,1))

outliers <- tso(timeSeries)
plot(outliers)

Arima of Tasking Priority 4 over 49 days (July9)

library(forecast)
true <- read.table("C:/Users/Rinor/Downloads/sgt/100.txt", quote="\"", comment.char="")
part <- read.table("C:/Users/Rinor/Downloads/sgt/70.txt", quote="\"", comment.char="")

timeSeries <- ts(true, frequency=1, start=c(0,1))
timeSeries2 <- ts(part, frequency=1, start=c(0,1))

fit <- auto.arima(timeSeries2)

plot(timeSeries, ylab="Tasking Performance", xlab="Day")

plot(forecast(fit, 14), ylab="Tasking Performance", xlab="Day")

Moving Average of Tasking Priority 4 over 49 days (July9)

mean <- rollmean(timeSeries,3)
ts.plot(timeSeries, mean, col=1:2, lwd=1:3, main="3pt Moving Average (Red) imposed onto Original (Black)", ylab="Tasking Perfomance", xlab="Day")

Correlation of Tasking Performance Priority 5 vs RCS (July 7)

july7cat5 <- read.csv("C:/Users/Rinor/Downloads/sgt/july7cat5.csv")
attach(july7cat5)
task <- Average.percentageTracked
rcs <- Average.measuredHitRCS
timeSeries3 <- ts(task, frequency=1, start=c(0,1))
timeSeries4 <- ts(rcs, frequency=1, start=c(0,1))
cor(timeSeries3, timeSeries4)
## [1] 0.8052314
library(clusterSim)
z1 <- data.Normalization(task,type="n1",normalization="column")
z2 <- data.Normalization(rcs,type="n1",normalization="column")
timeSeries9 <- ts(z1, frequency=1, start=c(0,1))
timeSeries10 <- ts(z2, frequency=1, start=c(0,1))
ts.plot(timeSeries9, timeSeries10, col=1:2, main="Tasking Perfomance Priority 5 (Black) vs RCS (Red)", xlab="Day", ylab="# of Std Dev from the Mean")

x <- unlist(task)
y <- unlist(rcs)
plot(x,y, pch = 16, cex = 1.3, col = "blue", main="Tasking Performance Priority 5 vs RCS",xlab="task", ylab="rrc")
abline(lm(y~x), col="red")

Correlation of Tasking Performance Priority 4 vs RCS (July 7)

july7cat4 <- read.csv("C:/Users/Rinor/Downloads/sgt/july7cat4.csv")
attach(july7cat4)
task2 <- Average.percentageTracked
rcs2 <- Average.measuredHitRCS
timeSeries5 <- ts(task2, frequency=1, start=c(0,1))
timeSeries6 <- ts(rcs2, frequency=1, start=c(0,1))
cor(timeSeries5, timeSeries6)
## [1] -0.4265296
ts.plot(timeSeries5, timeSeries6, col=1:2, main="Tasking Perfomance Priority 4 (Black) vs RCS (Red)", xlab="Day", ylab="Percentage Tracked or RCS")

x2 <- unlist(task2)
y2 <- unlist(rcs2)
plot(x2,y2, pch = 16, cex = 1.3, col = "blue", main="Tasking Performance Priority 4 vs RCS",xlab="task", ylab="rrc")
abline(lm(y2~x2), col="red")

Tasking 4 vs RCS 5 (July 7)

plot(x2,y, pch = 16, cex = 1.3, col = "blue", main="Tasking Performance Priority 4 vs RCS 5",xlab="task", ylab="rrc")
abline(lm(y~x2), col="red")

cor(x2,y)
## [1] 0.7802521

Correlation of Tasking Performance Priority 5 vs Radar Range Constant (July 9)

july9cat5 <- read.csv("C:/Users/Rinor/Downloads/sgt/july9cat5.csv")
attach(july9cat5)
task <- Average.percentageTracked
rrc <- Average.logr45_radar_range_constant
timeSeries7 <- ts(task, frequency=1, start=c(0,1))
timeSeries8 <- ts(rrc, frequency=1, start=c(0,1))
cor(timeSeries7, timeSeries8, use = "complete.obs")
## [1] 0.8377348
library(clusterSim)
z1 <- data.Normalization(task,type="n1",normalization="column")
z2 <- data.Normalization(rrc,type="n1",normalization="column")
timeSeries9 <- ts(z1, frequency=1, start=c(0,1))
timeSeries10 <- ts(z2, frequency=1, start=c(0,1))
ts.plot(timeSeries9, timeSeries10, col=1:2, main="Tasking Perfomance Priority 4 (Black) vs RRC (Red)", xlab="Day", ylab="# of Std Dev from the Mean")

x <- unlist(task)
y <- unlist(rrc)
plot(x,y, pch = 16, cex = 1.3, col = "blue", main="Tasking Performance Priority 5 vs RRC", xlab="task", ylab="rrc")
##abline(lm(y~x), col="red")

quad <- lm(y~poly(x,3))
x <- seq(.7,.9,.01)
pred <- predict(quad, list(x))
lines(x,pred,col="red",lwd=3)

Correlation of Tasking Performance Priority 4 vs Radar Range Constant (July 9)

july9cat4 <- read.csv("C:/Users/Rinor/Downloads/sgt/july9cat4.csv")
attach(july9cat4)
task <- Average.percentageTracked
rrc <- Average.logr45_radar_range_constant
timeSeries7 <- ts(task, frequency=1, start=c(0,1))
timeSeries8 <- ts(rrc, frequency=1, start=c(0,1))
cor(timeSeries7, timeSeries8, use = "complete.obs")
## [1] 0.9298924
library(clusterSim)
z1 <- data.Normalization(task,type="n1",normalization="column")
z2 <- data.Normalization(rrc,type="n1",normalization="column")
timeSeries9 <- ts(z1, frequency=1, start=c(0,1))
timeSeries10 <- ts(z2, frequency=1, start=c(0,1))
ts.plot(timeSeries9, timeSeries10, col=1:2, main="Tasking Perfomance Priority 4 (Black) vs RRC (Red)", xlab="Day", ylab="# of Std Dev from the Mean")

x <- unlist(task)
y <- unlist(rrc)
plot(x,y, pch = 16, cex = 1.3, col = "blue", main="Tasking Performance Priority 4 vs RRC",xlab="task", ylab="rrc")
#abline(lm(y~x), col="black")

quad <- lm(y~poly(x,4))
x <- seq(.3,.8,.01)
pred <- predict(quad, list(x))
lines(x,pred,col="red",lwd=3)

Elset Age Cat 4 vs Tasking

july9cat4elset <- read.csv("C:/Users/Rinor/Downloads/sgt/july9cat4elset")
attach(july9cat4elset)
x <- unlist(V1)
attach(july9cat4)
y <- unlist(Average.percentageTracked)
y <- y[1:49]

ryts <- decompose(ts(y, frequency = 7, c(0,1)))$random
xts <- ts(x, frequency = 7, c(0,1))

y <- as.numeric(ryts)
x <- as.numeric(xts)
plot(x,y)
abline(lm(y~x), col="red")

ts.plot(xts, ryts, ylab="elsetAge  /  taskingPercentage", xlab="Week", col=1:2, main="Elset Age (Black) and Tasking Performance without Trend (Red)")