Python kalman_filter Examples

Example #1

File: kalmanFilter.py Project: fedorayang/KF

 def train(self):
     """
     This fucntion will start the estimation. This is separated from addData.
     """
     beta = scipy.empty((self.t, self.n))
     b = self.initBeta
     V = self.initVariance
     Phi = self.Phi
     S = self.Sigma
     s = self.sigma
     y = self.respond.data
     X = self.regressors.data
     beta = kalman_filter(b, V, Phi, y, X, s, S)
     self.est = TimeSeriesFrame(beta, self.regressors.rheader,
                                self.regressors.cheader)
     return self

Example #2

File: kalmanFilter.py Project: idaohang/KF

 def train(self):
     """
     This fucntion will start the estimation. This is separated from addData.
     """
     beta = scipy.empty((self.t, self.n))
     b = self.initBeta
     V = self.initVariance
     Phi = self.Phi
     S = self.Sigma
     s = self.sigma
     y = self.respond.data
     X = self.regressors.data
     beta =  kalman_filter(b, V, Phi, y, X, s, S)
     self.est = TimeSeriesFrame(beta, 
                                self.regressors.rheader, 
                                self.regressors.cheader)
     return self

Example #3

n = 7
beta = scipy.empty(scipy.shape(X), dtype=numpy.float)
b = scipy.ones((n, 1), dtype=numpy.float) / float(n)
V = scipy.identity(n, dtype=numpy.float)
Phi = scipy.identity(n, dtype=numpy.float)
S = scipy.identity(n, dtype=numpy.float)
sigma = scipy.matrix([1.0], dtype=numpy.float)
Sigma = scipy.identity(n, dtype=numpy.float)
D = scipy.ones((1, n), dtype=numpy.float)
d = scipy.matrix(1., dtype=numpy.float)
G = scipy.identity(n, dtype=numpy.float)
a = scipy.zeros((n, 1), dtype=numpy.float)
c = scipy.ones((n, 1), dtype=numpy.float)
#        import code; code.interact(local=locals())

beta = clibregression.kalman_filter(b, V, Phi, y, X, sigma, Sigma, 1, D, d, G,
                                    a, c)

#(b, V) = kalman_predict(b,V,Phi, S)

# for i in xrange(len(X)):
#     beta[i] = scipy.array(b).T
#     (b,V, e,K) = kalman_upd(b,V, y[i] ,X[i], sigma, Sigma, 2, D, d,G,a,c)
#     (b, V) = kalman_predict(b,V,Phi, S)

# for i, (xs, ys) in enumerate(zip(X,y)):
#     beta[i,:] = scipy.array(b).T
#     (b,V, e,K) = kalman_upd(b,V, ys ,xs, sigma, Sigma, 2, D, d,G,a,c)
# #    print "b:\n", b
# #    print "V:\n", V
# #    print "e:\n", e
# #    print "K:\n", K

Example #4

File: test_kalmanFilter.py Project: idaohang/KF

n=7
beta = scipy.empty(scipy.shape(X), dtype = numpy.float)
b = scipy.ones((n,1), dtype = numpy.float)/float(n)
V = scipy.identity(n, dtype = numpy.float)
Phi = scipy.identity(n, dtype = numpy.float)
S = scipy.identity(n, dtype = numpy.float)
sigma = scipy.matrix([1.0], dtype = numpy.float)
Sigma = scipy.identity(n, dtype = numpy.float)
D = scipy.ones((1,n), dtype = numpy.float)
d = scipy.matrix(1., dtype = numpy.float)
G = scipy.identity(n, dtype = numpy.float)
a = scipy.zeros((n,1), dtype = numpy.float)
c = scipy.ones((n,1), dtype = numpy.float)
#        import code; code.interact(local=locals())

beta =  clibregression.kalman_filter(b, V, Phi,  y, X, sigma, Sigma, 1 ,D , d, G, a, c)

#(b, V) = kalman_predict(b,V,Phi, S)

# for i in xrange(len(X)):
#     beta[i] = scipy.array(b).T
#     (b,V, e,K) = kalman_upd(b,V, y[i] ,X[i], sigma, Sigma, 2, D, d,G,a,c)
#     (b, V) = kalman_predict(b,V,Phi, S)


# for i, (xs, ys) in enumerate(zip(X,y)):
#     beta[i,:] = scipy.array(b).T
#     (b,V, e,K) = kalman_upd(b,V, ys ,xs, sigma, Sigma, 2, D, d,G,a,c)
# #    print "b:\n", b
# #    print "V:\n", V
# #    print "e:\n", e