chrisconlan / automated_trading_with_r Goto Github PK

Hi, Chris, I am going through your book and downloaded the code from this repo when I run decisionGen.R I get the following:

source("decisionGen.R")
Error in quantile.default(v[2, (cols + 1):(2 * cols)], shThresh, na.rm = TRUE) :
'probs' outside [0,1]

I was running Rscipt plan.R

Can someone please post a sample of a DATA zoo object? I am using Tiingo to get stock data information and having issues to unlist the stocks. Because Yahoo API is not working any longer, I have issues replicating.

Any sample of a couple of stocks will work.

Thanks,

@chrisconlan

As Yahoo YQL is not working. It is possible to get some data from Quandl

Can somebody put some samples on how data downloaded from yahoo looks like or how will DATA frame should be looking at the end? Quandl has many ways to download data, one is with zoo package. So I am working on the code if it is useful for someone.

Hi Chris,

Trying out the example and had to comment line 1092 (https://github.com/chrisconlan/automated_trading_with_R/blob/master/Examples/cv-NelderMead-MACD.R#L1092).

This is unrelated and not sure if you've noticed or not, but the forum board trade-execution-money-management is full of spam (http://r.chrisconlan.com/index.php/forums/forum/strategy-backtesting-trading-research-and-automation/trade-execution-money-management/).

Thanks for the book. Enjoying it.

Can someone please guide me how to create legends like the ones on chapter 1? Seems like they are in two columns. Below the example from the book, my code and my result.

plot(y = Et, x = t, type = "l", col = 1, xlab = "Time", ylab = "Equity ($)", main = "Figure 1.3: Randomly Generated Equity Curves")
grid()
abline(h = 10000)
lines(y = Et2, x = t, col = 2)
lines(y = Eb, x = t, col = 8)
legend(x = "topleft", col = c(1, 2, 8), lty = c(1, 1), lwd = 2, ncol = 2, title = "NPMD   Burke",
    cex = 0.5,
    legend = c(
            paste0(round(NPMD_Et, 2)),
            paste0(round(NPMD_Et2, 2)),
            paste0(round(NPMD_Eb, 2)),
            paste0(round(Burke_Et, 2)),
            paste0(round(Burke_Et2,2)),
            paste0(round(Burke_Eb, 2))

            ))

I am getting the following output:

We need to modify 2-6 script and replace yahoo.R with quandl.R
What does this code exactly tries to do?

if(!weekend & !span){
      c <- as.numeric(substr(maxdate, start = 1, stop = 4))
      a <- as.numeric(substr(maxdate, start = 6, stop = 7)) - 1
      b <- as.numeric(substr(maxdate, start = 9, stop = 10))
      df <- yahoo(i, a = a, b = b, c = c)
      if(!is.null(df)){
        if(all(!is.na(df)) & nrow(df) > 0){
          df <- df[nrow(df):1]
          write.table(df, file = paste0(i, ".csv"), sep = ",",

We can add start and end dates to quandl_get function in Platform/functions/quandl.R as per below format from quandl.
data <- Quandl("FRED/GDP", start_date="2001-12-31", end_date="2005-12-31")

Do we need this?
Do we need start_date and end_date? or only start_date?