TAFT: Technical Analysis tools For Trading
version 1.0.2
Related documents: data
Important notice: Quotes and volumes (numpy-arrays given as arguments to the functions listed below) are required to be indexed so that the most old value has the biggest index while the most new one has index 0.
Indicators
AD-indicator
def ad( period=1, shift=0, hi=None, lo=None, cl=None, vo=None, prev=None )
period (int) - the period of the indicator, default: 1
shift (int) - the shift inside the data arrays (hi,lo,l,vo) to calculate the indicator at, default: 0
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
cl (numpy array, float) - CLOSE rates
vo (numpy array, float) - volumes
prev (float) - the value previously returned by the function, default: None
Returns (float) - the value of the indicator, 'None' if failed
ADX-indicator
def adx( period=14, shift=0, hi=None, lo=None, cl=None, prev=None )
period (int) - the period of the indicator, default: 14
shift (int) - the shift inside the data arrays (hi,lo,cl) to calculate the indicator for, default: 0
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
cl (numpy array, float) - CLOSE rates
prev (dict) - previously returned by the function, default: None
Returns (dict) - { 'adx': the ADX value, 'dx': the DX value, "+DI": the "+DI" value, "-DI": the "-DI value,
"+DMsm": the smoothed "+DM" value, "-DMsm": the smoothed "-DM" value, "TRsm": the smoothed 'true-range' value }, 'None' if failed
ATR - Average True Range
def atr( period=14, shift=0, hi=None, lo=None, cl=None, prev=None ):
period (int) - the period of the indicator, default: 14
shift (int) - the shift inside the data arrays (hi,lo,cl) to calculate the indicator for, default: 0
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
cl (numpy array, float) - CLOSE rates
prev (float) - the value previously returned by the function, default: None
Returns (dict) - { 'atr': the average true-range value, 'tr' - the last rue-range value }, 'None' if failed
Bollinger Bands
def bollinger( period=20, shift=0, nStds = 2.0, rates=None ):
period (int) - the period of the indicator, default: 20
shift (int) - the shift inside the 'rates' array to calculate the indicator for, default: 0
nStds (float) - the number of standard deviations to calculate 'upper' and 'lower' values of the indicator, default:2
rates (numpy array, float) - rates
Returns (dict) - { 'middle': the 'middle' value of the bollinger, 'upper': the upper value of the bollinger,
'lower' - the lower value of the bollinger, 'std': the standard deviation value ), 'None' if failed
CCI Commodity Channel Index indicator
def cci( period=20, shift=0, hi=None, lo=None, cl=None, cciConst=0.015 ):
period (int) - the period of the indicator, default: 20
shift (int) - the shift inside the data arrays (hi,lo,cl) to calculate the indicator for, default: 0
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
cl (numpy array, float) - CLOSE rates
cciConst (float) - the constant used in the indicator formula, default: 0.015
Returns (dict) - { 'cci': the value of the indicator, 'meanTypicalPrice': the mean 'typical price',
'meanDeviation': the mean deviation of typical price against the mean value }, 'None' if failed
EMA - Exponential Moving Average
def ema( period=10, shift=0, alpha=None, rates=None, prev=None ):
period (int) - the period of the indicator, default: 10
shift (int) - the shift inside the 'rates' array to calculate the indicator for, default: 0 (the last received rate)
alpha (float) - the smoothing factor, default: 2.0 / (period + 1.0)
rates (numpy array, float) - HIGH rates
prev (float) - the value previously returned by the function
Returns (float) - the EMA value, 'None' if failed
Stochastic-oscillator
def stochastic( periodK=14, periodD=3, shift=0, hi=None, lo=None, cl=None ):
periodK (int) - the period of the 'fast' stochastic line, default: 14
periodD (int) - the period of the 'slow' stochastic line, default: 3
shift (int) - the shift inside the data arrays (hi,lo,cl) to calculate the indicator for, default: 0
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
cl (numpy array, float) - CLOSE rates
Returns: the 'K' ('fast) stochastic value, 'D': the 'D' ('slow') tochastic value, 'None' if failed.
ROC - The Rate Of Change indicator
def roc( period=12, shift=0, rates=None ):
period (int) - the period of the indicator, default: 12
shift (int) - the shift inside the 'rates' array to calculate the indicator for, default: 0 (the last received rate)
rates (numpy array, float) - the rates
Returns (float) - the rate of change (ROC) value, 'None' if failed
RSI - Relative Strength Index
def rsi( period=14, shift=0, rates=None, prev=None ):
period (int) - the period of the indicator, default: 14
shift (int) - the shift inside the 'rates' array to calculate the indicator for, default: 0 (the last received rate)
rates (numpy array, float) - the rates
prev (dict) - the value previously returned by the function
Returns (dict) - { 'rsi':the RSI value, 'rs': the Relative Strength value,
'averageGain': the 'average gain' value, 'averageLoss': the 'average loss' value }, 'None' if failed
SMA - Simple Moving Average
def sma( period=10, shift=0, rates=None ):
period (int) - the period of the indicator, default: 10
shift (int) - the shift inside the 'rates' array to calculate the indicator for, default: 0 (the last received rate)
rates (numpy array, float) - the rates
Returns (float) - the SMA value, 'None' if failed
Simulation and Research
simulateTrade
Simulates trade - opening and closing position.
def simulateTrade( shift=0, hi=None, lo=None, tp=None, sl=None, tpSlides=False, slSlides=False, side=1, price=None, type=0 ):
shift (int) - the shift inside the 'rates' arrays (hi,lo,cl), i.e. the index at which the trade is 'opened'
hi (numpy array, float) - HIGH rates
lo (numpy array, float) - LOW rates
tp (float) - the 'take profit' value, if 'None' a huge value of (max(hi)-min(lo))*100.0 is used
sl (float) - the 'stop loss' value, if 'None' a huge value of (max(hi)-min(lo))*100.0 is used
tpSlides (boolean) - if 'True' a sliding take profit is used
slSlides (boolean) - if 'True' a sliding stop loss is used
side (int) - the value of '1' simulates open LONG, the value of '-1' simulates open short
price (float) - the 'current' (initial) price, if 'None' the value of 'current price' is the mean between lo[shift] and hi[shift]
Returns (dict): {
'profit' (float): the profit in points;
'closedAt' (int): the index value (inside the 'hi' and 'lo' arrays) where the trade was 'closed',
'-1' if neigther stop loss nor take profit has been reached till the end of the arrays;
'closePrice' (float): close price
}
Utilities
normalize
Normalizes the values of an array with the following formula: x[i] = (x[i] - ) /
def normalize( x, meanX=None, stdX=None, normInterval=[0,-1] ):
x (numpy array, float) - the array to be normalized
meanX (float) - the mean value to normalize with (if "None" the value will be calculated)
stdX (float) - the std value to normalize with (if "None" the value will be calculated)
normInterval (list of length 2, int) - the starting and the ending indexes inside the 'x' array.
The indexes denote a subarray used to calculate 'meanX' and 'stdX'.
if normInterval[1]==-1, the whole array is used.
Returns: nothing
readFinam
Loads rates from Finam file
def readFinam( fileName ):
fileName (string) - the name of the file with Finam rates.
Returns (dict): {
'op' (numpy-array, float): open rates, '0' index takes the latest rate
'hi' (numpy-array, float): high rates, '0' index takes the latest rate
'lo' (numpy-array, float): low rates, '0' index takes the latest rate
'cl' (numpy-array, float): close rates, '0' index takes the latest rate
'vol' (numpy-array, float): volumes, '0' index takes the volume of the latest period
'dtm' (list, datetime): date&time, '0' index takes the date and time of the latest period
}
