qoelet

# performing set difference between two vectors

order_grp_a <- sort(orders_table_a$orders_id)
order_grp_b <- sort(orders_table_b$orders_id)
order_diff <- setdiff(order_grp_b, order_grp_a)

qoelet

# load up libraries
library(ggplot2)
library(plyr)
library(XLConnect)

# try and read worksheet directly
original_data <- XLConnect::readWorksheet(loadWorkbook("/home/path/to/given_data.xlsx"), sheet=1)
"
Error: OutOfMemoryError (Java): GC overhead limit exceeded
"

qoelet

# Django templating style variable interpolation via https://github.com/jokergoo/GetoptLong

library(GetoptLong)

country <- "SG"
my_query <- "SELECT * FROM my_table WHERE country='{{country}}';"
my_query <- qq(my_query, code.pattern="\\{\\{CODE\\}\\}")
cat(my_query)
"
> cat(my_query)

qoelet

# Plotting a v-line on date

date_idx <- which(df$Date=="2013-11-20")

p <- ggplot(df, aes(Date, Value))
p + geom_line() + geom_vline(xintercept=as.numeric(df$Date[date_idx]))

qoelet

# load sample data (with duplicates reported earlier)

basket <- read.csv("/home/kenny/some_data_with_duplicates.csv", header=TRUE)

# remove duplicates and sort by column

basket <- unique(basket)
basket <- basket[with(basket, order(basket_id)), ] # sort by ident

qoelet

inspect(sort(basketApriori, by="confidence"))

"
  lhs    rhs                         support confidence lift
1 {}  => {Apparel and Accessories} 0.4846383  0.4846383    1
2 {}  => {Shoes}                   0.3488603  0.3488603    1
3 {}  => {Accessories}             0.1219029  0.1219029    1
"

# Lowering the support/confidence to negl levels 

qoelet

# Reload data from list via Quandl

watchSymbols <- c(
    "GOOG/NYSE_TWTR",
    "GOOG/NASDAQ_FB"
)

watchNames <- c(
    "twitter",
    "facebook"

qoelet

# reverse columns in a data frame 

reverse_df <- function(df) {
  col_names <- names(df)
  for(col_idx in 1:length(col_names)) {
    df[[col_idx]] <- rev(df[[col_idx]])
  }
  return(df)
}

qoelet

vix <- Quandl("YAHOO/INDEX_VIX")
boxPlot(vix$Close)
describe(vix$Close)

"
vix$Close 
      n missing  unique    Mean     .05     .10     .25     .50     .75     .90     .95 
   6031       0    2280   20.21   11.54   12.25   14.49   18.53   23.69   29.73   34.63 

lowest :  9.31  9.48  9.70  9.82  9.89, highest: 72.67 74.26 79.13 80.06 80.86 

qoelet

library("RPostgreSQL")

drv <- dbDriver("PostgreSQL")
con <- dbConnect(drv, dbname="remote_db", user="kenny", port="5432", host="remote_db_host")

# create an environment container
myData <- new.env()

# fetch data
cur <- dbSendQuery(con, "select * from some_table order by date desc limit 1000;")