Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

c0stern1 - callbull - case - cdff - changetype - choosevariableX - cobfidenceb - confidencey - cos - covcheck - creal - csqrt - cvdec
c0stern1
caanalyse supplementary macro for corresp which performs correspondence analysis for a contingency table
cabs Absolute value of a complex array
cacheckvar supplementary macro for corresp which performs correspondence analysis for a contingency table
cadd Complex addition of two arrays
cadisplay supplementary macro for corresp which performs correspondence analysis for a contingency table
cagetdata supplementary macro for corresp which performs correspondence analysis for a contingency table
cagetdata supplementary macro for corresp which performs correspondence analysis for a contingency table
calibrIC Auxiliary routine for rICfil Calibrates the robust IC's for a given State Space model to a given relative efficiency loss in terms of the MSE in the ideal model. The state-space model is assumed to be in the following form:

y_t = H x_t + v_t

x_t = F x_t-1 + w_t

x_0 ~ (mu,Sig), v_t ~ (0,Q), w_t ~ (0,R)

All parameters are assumed known.

calibrLS Auxiliary routine for rLSfil Calibrates the robust LS- Filter for a given State Space model to a given relative efficiency loss in terms of the MSE in the ideal model. The state-space model is assumed to be in the following form:

y_t = H x_t + v_t

x_t = F x_t-1 + w_t

x_0 ~ (mu,Sig), v_t ~ (0,Q), w_t ~ (0,R)

All parameters are assumed known.

callbull calculates the results of a Bull Call Spread for the context of option pricing
canaxe supplementary macro for corresp which performs correspondence analysis for a contingency table
canbw does the canonical bandwith transformation of a bandwith value of kernel K1 into an equivalent bandwidth for Kernel K2.
canegat supplementary macro for corresp which performs correspondence analysis for a contingency table
canker does the canonical bandwith transformation of a bandwith value of kernel K1 into an equivalent bandwidth for Kernel K2.
cartcv Performs cross validation for the CART: subtracts from the data in a given number of ways a test set, with the rest of the data a regression tree is formed and a sequence of subtrees is pruned from the initial tree. For each tree, the test set is used to calculate the prediction error.
cartregr applies the the tree to a user given data set.
cartsplit Computes a regression tree.
cartsplitopt sets optional parameters for cartsplit (spliting of for classification and regression trees)
cartsplitout
case Inside a switch-endsw block case controls the execution of an alternative. If the condition of case is true, the following block is executed similar to an if-endif statement. The keyword break serves as end marker of case and leaves the switch block at the position of endsw. When break is omitted, the next consecutive case is processed. If the program's counter comes to default, the following block is executed in any case.

castr supplementary macro for corresp which performs correspondence analysis for a contingency table
categorize creates dummy variables from a data with respect to distinct realizations. The default reference category is the minimal value in each column. Alternatively, categorization can be done by giving a value or the index (rank among the realizations) in a column.
cceil Computes ceil for a complex array
cconj Conjugated array
ccos Complex cosine
ccosh Complex hyperbolic cosine
cdfb Returns the values of the beta-distribution function with parameters a and b for the elements of an array.
cdfbin computes the cumulative distribution function of a binomial distribution
cdfc Returns the values of the chi-quare distribution function with d degrees of freedom for the elements of an array.
cdff Returns the values of the F-distribution function with d1 and d2 degrees of freedom for the elements of an array.
cdfn Returns the values of the standard normal distribution function for the elements of an array.
cdft Returns the values of the t-distribution function with d degrees of freedom for the elements of an array.
cdfx cdfx returns the value of the extreme value and generalized Pareto distribution functions for elements of a vector.
cdiv Complex division
ceil Returns the smallest integer value greater or equal to each element of an array.
cexp Complex exponential
cfc1diff Forecasting undifferenced time series in VAR models
cfloor Computes floor for a complex array
changename changes the names of selected variables
changetype changes the types of selected variables and shows if they excluded or included
char Convert numbers to ASCII characters in strings.
chbase Changes interactively the wavelet coeffients. You may choose between Haar, Daubechies2 and Coiflet2.
chfunc Generates specific functions (Jump, Up-down, Sine, Freq. sine and Doppler). If all entries of sel are zero then you can choose interactively the function otherwise the selected function will be generated.
chol computes the Cholesky decomposition of a symmetric, positive definite matrix
chold The function chold is calculating the triangularisation and the Cholesky decomposition of x into matrices b and d, so that b'*d*b = x.
choosegroup selction of group variables (discrete type)
choosevariable selction of variables
choosevariable2 selction of variables to transform
choosevariablep selction of variables in regression context
choosevariableX selction of X variables in regression context
chview Enforces a specific view of the wavelet mother coefficients (Standard, Ordered, Circle and Partial sum). If none is selected then the old view will be returned.
ciirboot Computes two sided bootstrap confidence intervals for impulse responses for a K-dimensional VAR(p) by resampling the estimated residuals. The confidence intervals are computed using the methodology of Hall (The Bootstrap and the Edgeworth Expansion, 1992) and Efron & Tibshirani (An Introduction to the Bootstrap, 1993).
cimag Extracts imaginary part of a complex array
cinv Complex inverse matrix
cir displays the yield curve for given parameters under the model of Cox/Ingersoll/Ross (1985)
cln Complex natural logarithm
cmatdiv Computes the complex solution of A x = b
cmatmul Computes the complex matrix multiplication of X and Y
cmul Complex multiplication
cobfidenceb computes the 95% confidence intervals for each beta after regression analysis
coeffba auxiliary quantlet for full VAR model analysis
coeffest estimates the coefficients of a full VAR model
coeffss estimates parameters of Subset VAR
collinearity performs a collineatity diagnostic after regression analysis and shows the eigenvalues and condition indices of the independent variables
colorcube Displays a multi-color cube
cols Returns the number of columns in an array.
committee This macro computes a committee of networks with nets of the form single layer feedforward perceptron. The macro can be used alone or in connection with the library ISTA. The standalone version also needs the parameter data. Just choose 0 for the input. The number of nets to build the committee can be chosen. The data will be splitted with this number to build the different datasets. The weight for the cases for the training of the net can be chosen, the numbers of hidden units and additional information concerning the weights of the units. Different optional parameters can be chosen to build the architektur of the network. The choice holds for every single net. The default values are chosen in order to solve a linear regression problem. The optional parameters constits of 8 values. Boolean values for linear output, entropy error function, log probability models and for skip connections (direkt links). The fifth values is the maximum value for the starting weights, the sixth is the weight decay, the seventh the maximum number of iterations and the the last value generates the output concerning the architekur of the net if it is equal to one. The output consits of the Error and MSE of the single nets and for all cases. Additionally the R^2 for the average of the nets and the R^2 of the committee are shown.

comp Checks whether an object has a specific component or not. If the first argument is a string, the object with the specified name is regarded as a list object.
complex Generates a complex array
confidencey computes the 95% confidence intervals for the unstandardized predicted values after regression analysis
conting crosses two categorical variables (for instance partitions from cluster analysis) and builds up contingency table
contmax computes a linkage table between the rows and columns of a contingency table by maximum value of correspondence. The number of correspondences is the minimum of number or dimensions of the contingency table.
contour2 contour2 computes lines and points for a contourplot of a three dimensional dataset at level c
contour3 contour3 computes lines and points for a contourplot of a four dimensional dataset at level c
conv conv performs the convolution of a step kernel function and a function over a p-dimensional equidistant grid.

cor2dist transforms the values of the upper triangle of a correlation matrix into distances, and it stores these distances into a vector regarding the sequence described in agglom
corr Computes the correlation (Bravais-Pearson) structure of a given array.
corresp corresp executes Correspondence Analysis which analyses and describes a contingency table cross-tabulations) in terms of a reduced number of dimensions. Correspondence Analysis can be viewed as finding the best simultaneous representation of two sets that comprise the rows and columns of a data matrix, in order to obtain a summary description for large tables (cross-tabulations). This technique can be helpful in finding important underlying characteristics which might not be directly observed in the data. Graphical visualizations provide an insight and understanding tool for interpreting the data.
corrint computes the correlation integral for time series
cos Returns the cosine in radian of the elements of an array.
cosh Returns the hyperbolic cosine of the elements of an array.
cosi cosi computes the cosine kernel, multivariate
countNaN Counts how many missing values (NaN) are in an array.
countNotNumber Counts how many elements of an array are missing values, infinity or -infinity (NaN,Inf or -Inf).
cov Computes the covariance structure of a given array.
covabc Covariance matrix of C=A*B, Reduced Rank VAR Model
covabrr covariance matrix A*B, reduced rank VAR model
covarr covariance matrix A, reduced rank VAR model
covbrr Covariance matrix B, reduced rank VAR Model
covcheck checks if the covariance matrix is singular
covfore2 Computes forecast MSE matrix for undiff. time series
covforec calculates the forecast MSE matrix for VAR models
covmatrix computes the covariance matrix for beta after regression analysis
covmlrr covariance matrix used for reduced rank models
covmwgen generates covariance matrix of the mean in VAR
covres auxiliary quantlet for full VAR models
cplot Plot of x and y in absolute space
cplotfunc Plots a function of x and y-space
cpolar Complex numbers in polar coordinates
creal Extracts real part of an complex array
createcolor createcolor allocates the colors for user

createdisplay createdisplay create a display for further plotting the datasets or texts.

createportnumber
criterss calculates selection criteria for Subset VAR
crosstable computes pairwise crosstables from all columns of a data matrix, gives the result of a Chi-square independence test and computes contingency coefficients.
csin Complex sine
csinh Complex sine hyperbolicus
csort csort sorts the rows of a complex matrix with respect to the absolute value of the complex numbers. If a column c is specified the rows of the matrix will be ordered with respect to the elements of column c in ascending (descending) order.
csortcol sorts with respect to either a real part of a column or an imaginary part of a column c. If 1 <= c <= cols(xr) it sorts after the real part of x, if cols(xr) < c <= cols(xr)+cols(xi) it sorts the imaginary part after column c-cols(xr).
csqrt Complex squareroot
csub Complex subtraction two arrays of complex numbers
ctan Complex tangens
ctanh Complex tangens hyperbolicus
cumprod cumprod computes the cumulative product of the elements in an array regarding a given dimension.
cumsum cumsum computes the cumulative sum of the elements an array regarding a given dimension.
cv runs a cross validation over the hidden units
cvdec runs a cross validation over the weight decay

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

(C) MD*TECH Method and Data Technologies, 17.8.2000