maureenmao / sec-price

Sec-price

#Name the data as montly_sec_only

Q1

Check the missing value

    sum(is.na(c(monthly_sec_only$Partyname, monthly_sec_only$Bmunit)))

Import functions

    library(dplyr)
    library(tidyr)
    library(reshape2)
    library(fBasics)
    library(stats)
    library(car)

Kick out the duplicate data

    t <- select(monthly_sec_only, Partyname, Bmunit)
    t1 <- t[!duplicated(t),]

Arrange and sort

    arrange(t1, Partyname)
    num <- table(t1$Partyname)
    sort(num, decreasing=TRUE)

Maximum is the first one in the list, "E.ON UK plc 21"

Q2

    p <- data.frame(monthly_sec_only$Month , monthly_sec_only$B_Sec, monthly_sec_only$Sec_Mwh, stringsAsFactors = FALSE)
    p <- arrange(p, monthly_sec_only$Month)

Check NAs

    p0 <- na.omit(p)
    dim(p)
    dim(p0)
    #There are 10 missing values
    list <- which(rowSums(is.na(p)) !=0)
    p_na <- p[list, ]
    p_na

In this case we simply wipe NAs out because the missing data has little influence on the result

Define the montly price

    names(p0)
    p0 <- rename (p0, Month = "monthly_sec_only.Month", Bid = "monthly_sec_only.B_Sec", Volume = "monthly_sec_only.Sec_Mwh")
    names(p0)
    Month <- group_by(p0, Month)
    s <- sum(p0[ ,2])
    sec_price <- summarize(Month, secprice = sum(Bid * Volume / s))

Create a time series data

    price <- ts(sec_price$secprice, frequency = 12 , start = c(2005,11))
    plot(price, xlab = 'Year', ylab = '(Secondary)Volume Weighted Average Price', main = 'Monthly Price Line', type = 'l')

Q3

    ##Create a new chart of pence per KWh of coal, oil and gas from chart "Quarterly" and extract column A,B,D,F,G
    ##Impoer the refined data from another working sheet called "input price"
    coal_gas_prices <- read_excel("D:/JHU/Career/1/coal_gas_prices.xls", sheet = "Input Price")
    View(coal_gas_prices)
    names(coal_gas_prices)

    input_prices <- coal_gas_prices[ ,3:5]
    head(input_prices)

Merge the data to a dataset

Modify the length of second price to the length of quarterly data in the sheet called "Sec"

    sec <- read_excel("D:/JHU/Career/1/coal_gas_prices.xls", sheet = "Sec")

Create join variable "qt"

    sec$qt <- rep(1:16, each = 3)
    input_prices$qt <- rownames(input_prices)

    sec3 <- merge(sec, input_prices)
    sec3 <- select(sec3,secprice:Gas)
    sec3 <- sec3[-1, ]

Check the merged dataset

    head(sec3)
    View(sec_price)

Calling lm

    fit <- lm (secprice ~ . , data=sec3)
    summary (fit) $coefficient

Interpretation

#R-square equals to 0.1195 and adjusted R-square equals to 0.05804, which is acceptable #The probility tells us that the price of caol, oil and gas all have significant statistical impacts on the level of 5% #Which means a $0.5 decrease in the coal price would lead to a $1 rise in the second price; a $0.03, $0.1 increase in the oil and gas price could lead to a $1 rise in the second price

Verify the multicollinearity

    vif(fit)

#VIf less than 10, which means there is no multicollinearity

From the analysis, we can conclude that although oil price has a little impact on the second price, caol price could have an adverse impact.

So in order to analysis the second price of the plants, we should also pay close attention to the price of the coal market.