private boolean paramInBounds(Edge edge, double newValue) {
edgeParameters.put(edge, newValue);
Map<Node, Double> errorVariances = new HashMap<Node, Double>();
for (Node node : semPm.getVariableNodes()) {
Node error = semGraph.getExogenous(node);
double d2 = calculateErrorVarianceFromParams(error);
if (Double.isNaN(d2)) {
return false;
}
errorVariances.put(error, d2);
}
if (!MatrixUtils.isPositiveDefinite(errCovar(errorVariances))) {
return false;
}
return true;
}
public GeneralizedSemPm(SemPm semPm) {
this(semPm.getGraph());
// Write down equations.
try {
List<Node> variableNodes = getVariableNodes();
for (int i = 0; i < variableNodes.size(); i++) {
Node node = variableNodes.get(i);
List<Node> parents = getVariableParents(node);
StringBuilder buf = new StringBuilder();
for (int j = 0; j < parents.size(); j++) {
if (!(variableNodes.contains(parents.get(j)))) {
continue;
}
Node parent = parents.get(j);
Parameter _parameter = semPm.getParameter(parent, node);
String parameter = _parameter.getName();
Set<Node> nodes = new HashSet<>();
nodes.add(node);
referencedParameters.put(parameter, nodes);
buf.append(parameter);
buf.append("*");
buf.append(parents.get(j).getName());
setParameterExpression(parameter, "Split(-1.5, -.5, .5, 1.5)");
setStartsWithParametersTemplate(parameter.substring(0, 1), "Split(-1.5, -.5, .5, 1.5)");
setStartsWithParametersEstimationInitializaationTemplate(
parameter.substring(0, 1), "Split(-1.5, -.5, .5, 1.5)");
if (j < parents.size() - 1) {
buf.append(" + ");
}
}
if (buf.toString().trim().length() != 0) {
buf.append(" + ");
}
buf.append(errorNodes.get(i));
setNodeExpression(node, buf.toString());
}
for (Node node : variableNodes) {
Parameter _parameter = semPm.getParameter(node, node);
String parameter = _parameter.getName();
String distributionFormula = "N(0," + parameter + ")";
setNodeExpression(getErrorNode(node), distributionFormula);
setParameterExpression(parameter, "U(0, 1)");
setStartsWithParametersTemplate(parameter.substring(0, 1), "U(0, 1)");
setStartsWithParametersEstimationInitializaationTemplate(
parameter.substring(0, 1), "U(0, 1)");
}
variableNames = new ArrayList<>();
for (Node _node : variableNodes) variableNames.add(_node.getName());
for (Node _node : errorNodes) variableNames.add(_node.getName());
} catch (ParseException e) {
throw new IllegalStateException("Parse error in constructing initial model.", e);
}
}
/** @return the list of variable nodes of the model, in order. */
public List<Node> getVariableNodes() {
return semPm.getVariableNodes();
}
/**
* @return For compatibility only. Returns the variable means of the model. These are always zero,
* since this is a standardized model. THESE ARE ALSO NOT PARAMETERS OF THE MODEL. ONLY THE
* COEFFICIENTS ARE PARAMETERS.
*/
public double[] means() {
return new double[semPm.getVariableNodes().size()];
}
/**
* Constructs a new standardized SEM IM from the freeParameters in the given SEM IM.
*
* @param im Stop asking me for these things! The given SEM IM!!!
* @param initialization CALCULATE_FROM_SEM if the initial values will be calculated from the
* given SEM IM; INITIALIZE_FROM_DATA if data will be simulated from the given SEM,
* standardized, and estimated.
*/
public StandardizedSemIm(SemIm im, Initialization initialization) {
this.semPm = new SemPm(im.getSemPm());
this.semGraph = new SemGraph(semPm.getGraph());
semGraph.setShowErrorTerms(true);
if (semGraph.existsDirectedCycle()) {
throw new IllegalArgumentException("The cyclic case is not handled.");
}
if (initialization == Initialization.CALCULATE_FROM_SEM) {
// This code calculates the new coefficients directly from the old ones.
edgeParameters = new HashMap<Edge, Double>();
List<Node> nodes = im.getVariableNodes();
TetradMatrix impliedCovar = im.getImplCovar(true);
for (Parameter parameter : im.getSemPm().getParameters()) {
if (parameter.getType() == ParamType.COEF) {
Node a = parameter.getNodeA();
Node b = parameter.getNodeB();
int aindex = nodes.indexOf(a);
int bindex = nodes.indexOf(b);
double vara = impliedCovar.get(aindex, aindex);
double stda = Math.sqrt(vara);
double varb = impliedCovar.get(bindex, bindex);
double stdb = Math.sqrt(varb);
double oldCoef = im.getEdgeCoef(a, b);
double newCoef = (stda / stdb) * oldCoef;
edgeParameters.put(Edges.directedEdge(a, b), newCoef);
} else if (parameter.getType() == ParamType.COVAR) {
Node a = parameter.getNodeA();
Node b = parameter.getNodeB();
Node exoa = semGraph.getExogenous(a);
Node exob = semGraph.getExogenous(b);
double covar = im.getErrCovar(a, b) / Math.sqrt(im.getErrVar(a) * im.getErrVar(b));
edgeParameters.put(Edges.bidirectedEdge(exoa, exob), covar);
}
}
} else {
// This code estimates the new coefficients from simulated data from the old model.
DataSet dataSet = im.simulateData(1000, false);
TetradMatrix _dataSet = dataSet.getDoubleData();
_dataSet = DataUtils.standardizeData(_dataSet);
DataSet dataSetStandardized = ColtDataSet.makeData(dataSet.getVariables(), _dataSet);
SemEstimator estimator = new SemEstimator(dataSetStandardized, im.getSemPm());
SemIm imStandardized = estimator.estimate();
edgeParameters = new HashMap<Edge, Double>();
for (Parameter parameter : imStandardized.getSemPm().getParameters()) {
if (parameter.getType() == ParamType.COEF) {
Node a = parameter.getNodeA();
Node b = parameter.getNodeB();
double coef = imStandardized.getEdgeCoef(a, b);
edgeParameters.put(Edges.directedEdge(a, b), coef);
} else if (parameter.getType() == ParamType.COVAR) {
Node a = parameter.getNodeA();
Node b = parameter.getNodeB();
Node exoa = semGraph.getExogenous(a);
Node exob = semGraph.getExogenous(b);
double covar = -im.getErrCovar(a, b) / Math.sqrt(im.getErrVar(a) * im.getErrVar(b));
edgeParameters.put(Edges.bidirectedEdge(exoa, exob), covar);
}
}
}
this.measuredNodes = Collections.unmodifiableList(semPm.getMeasuredNodes());
}