WarpingDistance
WarpingCorrespondence
CanonicalWarpingDistance
CanonicalWarpingCorrespondence
ImageDisplacements
DynamicImage
FeatureExtraction
ComponentMeasurements
SelectComponents
ArrayResample
ResamplingAlgorithmData
BinaryDistance
SequenceAlignment

Finance & Economics

Compare and Predict Stock Prices

Use dynamic time warping to predict stock prices for the next 30 days

Open in Cloud
Download Notebook

Use

FinancialData

to obtain the HPQ stock prices for the first quarter of 2017:

In[1]:=

recent=QuantityMagnitude[FinancialData["HPQ",{{2017,1,1},{2017,3,31}}]["Values"]];

Consider the historical data fro 2010 to 2016. Obtain the values and the dates separately:

In[2]:=

history=FinancialData["HPQ",{{2010,1,1},{2016,12,31}},"DateValue"];{histDates,hist}={history["Dates"],QuantityMagnitude[history["Values"]]};

Find the best matching subsequence of historical data, using

WarpingCorrespondence

In[3]:=

{corrHist,corrRecent}=WarpingCorrespondence[hist,recent,Method{"MatchingInterval""Flexible"}];

Detect the historical interval most similar to quarter one of 2022:

In[4]:=

{m,n}=corrHist〚{1,-1}〛;histDates〚{m,n}〛

Out[4]=



Mon 12 May 2014 00:00:00GMT-5

Wed 3 Sep 2014 00:00:00GMT-5



Visually compare recent data and the best historical match:

In[5]:=

DateListPlot{AssociationThread[Take[histDates〚m;;n〛,UpTo@Length[recent]],Take[recent,UpTo[n-m+1]]],AssociationThread[histDates〚m;;n〛,hist〚m;;n〛]},



Out[5]=

	2022 data
	Best historical match

Predict the stock prices for the next 30 days based on historical data:

In[6]:=

l=Length[recent];colDat=ColorData[97];offset=Last[recent]-hist〚n〛;hist30d=hist〚n;;n+30〛;ListLinePlot{recent,{Length[recent]+Range[31],hist30d+Last[recent]-hist〚n〛},hist〚m;;n〛,{n-m+Range[31],hist30d}},



Out[6]=

2017 data

30 days prediction

Best historical match

Next 30 days of historical data

Related Symbols

WarpingDistance
WarpingCorrespondence
CanonicalWarpingDistance
CanonicalWarpingCorrespondence
ImageDisplacements
DynamicImage
FeatureExtraction
ComponentMeasurements
SelectComponents
ArrayResample
ResamplingAlgorithmData
BinaryDistance
SequenceAlignment

Publisher Information

Contributed by: Wolfram Staff