Back-testing Trades

#*****************************************************************
# Load signals from http://www.decisionmoose.com/Moosistory.html
#*****************************************************************
library(SIT)
load.packages('quantmod')


filename = 'data/decisionmoose.Rdata'
if(!file.exists(filename)) {
	url = 'http://www.decisionmoose.com/Moosistory.html'
	txt = join(readLines(url))

	# extract transaction history
	temp = extract.table.from.webpage(txt, 'Transaction History', has.header = F)
		temp = trim(temp[-1,2:5])
 	colnames(temp) = spl('id,date,name,equity')
 
 id = as.numeric(temp[,'id'])
 	index = ifna(id > 0,F)
 temp = temp[index,]

 temp[,'equity'] = gsub('\\.','',gsub(',','',gsub('\\$','',temp[,'equity'])))
 info = data.frame(stringsAsFactors = F,
 id = as.numeric(temp[,'id']), 
 equity = as.numeric(temp[,'equity']),
 date = as.Date(temp[,'date'],'%m.%d.%Y'),
 tickers = toupper(trim(gsub('\\)','', sapply(temp[,'name'], spl, '\\('))))[2,]
 )
 info = info[!is.na(info$date),]
save(info, file=filename)
}

load(file=filename)
 
 
#plota(make.xts(info$equity, info$date), type='l')
 
#*****************************************************************
# Load historical data
#*****************************************************************
tickers = unique(info$tickers)

# load saved Proxies Raw Data, data.proxy.raw
load('data/data.proxy.raw.Rdata')

# define Cash (3moT) 3MOT = BIL+TB3M 
tickers = gsub('3MOT','3MOT=BIL+TB3M', tickers)

#US:SAF Scudder New Asia Fund (SAF), Merged into DWS Emerging Markets Equity Fund
#https://www.backrecord.com/topic/us-saf
tickers = gsub('SAF','SAF=SEKCX', tickers)


# add dummy stock, to keep dates of trades, if they do not line up with data
dummy = make.stock.xts(make.xts(info$equity, info$date))

data <- new.env()
getSymbols.extra(tickers, src = 'yahoo', from = '1970-01-01', env = data, raw.data = data.proxy.raw, auto.assign = T)

	# optionally investigate splits not captured by Adjusted
	#data.clean(data, min.ratio=3)

  for(i in ls(data)) data[[i]] = adjustOHLC(data[[i]], use.Adjusted=T)
  
  #print(bt.start.dates(data))
  
  data$dummy = dummy

bt.prep(data, align='keep.all', dates='2000::',fill.gaps = T)

bt.prep.remove.symbols(data, 'dummy')

#plota.matplot(scale.one(data$prices),main='Asset Perfromance')

#*****************************************************************
# Setup
#*****************************************************************
prices = data$prices

models = list()

#*****************************************************************
# Code Strategies, SPY - Buy & Hold
#*****************************************************************
data$weight[] = NA
	data$weight$SPY = 1
models$SPY = bt.run.share(data, clean.signal=T, silent=T)

#*****************************************************************
# Create weights
#*****************************************************************
weight = NA * prices

for(t in 1:nrow(info)) {
	weight[info$date[t],] = 0
 weight[info$date[t], info$ticker[t]] = 1
}

data$weight[] = 	weight
models$decisionmoose = bt.run.share(data, clean.signal=F, trade.summary=T, silent=T)#, do.lag = 2)


#*****************************************************************
# Create Report
#****************************************************************** 
plota.matplot(scale.one(data$prices),main='Asset Perfromance')

#strategy.performance.snapshoot(models, title = 'decisionmoose', data = data)
plotbt(models, plotX = T, log = 'y', LeftMargin = 3, main = NULL)	    	
	mtext('Cumulative Performance', side = 2, line = 1)

print(plotbt.strategy.sidebyside(models, make.plot=F, return.table=T))

m = 'decisionmoose'
plotbt.transition.map(models[[m]]$weight, name=m)
		legend('topright', legend = m, bty = 'n')

print(last.trades(models$decisionmoose, smain='decisionmoose',make.plot=F, return.table=T))