modelbase.oxh

Modelbase : Database : Sample class reference

Modelbase function members (public by default)
virtual	Batch(const sBatch, ...)
	Runs batch code (OxModel only).
virtual	BatchMethod(const sMethod)
	Used to set the method from batch (OxModel only).
virtual	Buffering(const bBufferOn)
	Switches output buffering on and off (when using OxPack).
virtual	ClearEstimation()
	Clears estimation status.
virtual	ClearModel()
	Clears the estimated model and calls `Modelbase::ClearEstimation`.
virtual	Covar()
	Computes the covariance matrix.
	DbDrawTMatrix(const iArea, const mYt, const asY, const iT1, ...)
	Like DrawTMatrix, but using the underlying database sample information.
	DeSelect()
	Remove the current variable and sample selection.
	DeSelectByIndex(const iSel)
	Removes variable(s) from the selection.
	DeSelectByName(const sVar, const theGroup, const iLag)
	Removes variable(s) from the selection.
virtual	DoEstimateDlg(const iFirstMethod, const cMethods, const sMethods, const bForcAllowed, const bRecAllowed, const bMaxDlgAllowed)
	Create and execute an OxPack dialog for model Estimation.
virtual	DoEstimation( vParFree)
	Do the model estimation, called from `Modelbase::Estimate`.
virtual	DoFormulateDlg(const iLagMode, ...)
	Call the OxPack model Formulation dialog using
virtual	DoOption(const sOpt, const val)
	Implement an option as requested by `Modelbase::DoOptionsDlg`.
virtual	DoOptionsDlg(const aMoreOptions)
	Create and execute an OxPack dialog for model options;
virtual	DoSettingsDlg()
	Call the OxPack model Settings dialog.
virtual	Estimate()
	Prepare and estimate the model.
virtual	FindGroup(const theGroup)
	Finds the group identifier.
virtual	FindMethod(const theMethod)
	Finds the method identifier.
	FixPar(const iP, const dFix)
	Fixes parameters.
	ForceYlag(const iYgroup)
	Reclassifies lagged iYgroup (usually Y_VAR) variables
	FreePar(const iP)
	Frees parameters.
virtual	GetBatchCode()
	Gets the Batch code for the entire model.
virtual	GetBatchDatabase()
	Creates the database text for batch (OxModel only).
	GetBatchEquation(const sY, const asX)
	Creates the equation formulation text for batch (OxModel only).
virtual	GetBatchEstimate()
	Creates the model estimation text for batch (OxModel only).
virtual	GetBatchModel()
	Creates the model formulation text for batch (OxModel only).
virtual	GetBatchModelSettings()
	Creates the model settings text for batch (OxModel only).
virtual	GetBatchModule(const sModule, const sPackage)
	Creates the module text for batch (OxModel only).
	GetCovar()
	Returns covariance matrix.
	GetCovarRobust()
	Returns robust covariance matrix.
	GetForecastData(const iGroup, const mnLag, const mxLag, const cTforc, ...)
	Gets data from the specified model for forecasting.
	GetFreePar()
	Get the vector of free parameters.
	GetFreeParCount()
	Gets number of free parameters p.
	GetFreeParNames()
	Get the names of free parameters.
virtual	GetGroupLabels()
	Defines the Y_VAR,X_VAR,... constants as array of strings.
	GetInEstimationSample(const iT1=-1, const iT2=-1)
	Gets the subsample text relative to the estimation sample.
	GetInForecastSample(const iT1, const iT2=-1)
	Gets the subsample text relative to the estimation sample.
	GetLogLik()
	Gets the loglikelihood value.
virtual	GetLongRunInfo()
	Get the info of longrun parameters.
virtual	GetLongRunNames()
	Get the names of longrun parameters.
	GetMethod()
	Gets the estimation method.
virtual	GetMethodLabel()
	Gets the text label of the estimation method.
virtual	GetMethodLabels()
	Defines the M_NONE,M_OLS,... constants as array of strings.
virtual	GetModelLabel()
	Gets the label of the model.
virtual	GetModelSettings()
	Called by OxPack to get the model estimation settings (by default
	GetModelStatus()
	Gets the model status.
virtual	GetOxCode()
	Gets the Ox code for the entire model.
virtual	GetOxDatabase(const sObj, const sClass)
	Gets the Ox code to load the database.
virtual	GetOxDecl(const sObj, const sClass)
	Gets the Ox code to declare the modelling object.
virtual	GetOxEstimate(const sObj, const sClass)
	Gets the Ox code for the model estimation.
virtual	GetOxModel(const sObj, const sClass)
	Gets the Ox code for the model specification.
virtual	GetOxModelSettings(const sObj, const sClass)
	Gets the Ox code for the model settings.
virtual	GetPackageName()
	Returns package name.
virtual	GetPackageVersion()
	Returns package version.
	GetPar()
	Get the vector of all parameters.
	GetParCount()
	Gets get number of parameters q (including fixed).
virtual	GetParNames()
	Get the names of all parameters.
	GetParStatus()
	Returns full parameter info.
virtual	GetParTypes()
	Returns array of type letters for each model variable.
	GetPrint()
	Gets the current print status.
virtual	GetResVar()
	Returns residual variance.
virtual	GetResiduals()
	Gets the residuals.
	GetResult()
	Gets the result of estimation.
	GetStdErr()
	Returns the standard errors of all parameters (0 for fixed).
	GetStdErrRobust()
	Returns the robust std.errors of all parameters (0 for fixed).
	GetX()
	Returns the X matrix.
	GetY()
	Returns the Y matrix.
virtual	GetcDfLoss()
	Returns degrees of freedom lost (number of free parameters).
virtual	GetcT()
	Returns actual no of observations used in the estimation.
	GetcX()
	Returns number of X variables.
	GetcY()
	Returns number of Y (dependent) variables.
	GetcYlag()
	Returns number of columns in m_mX that are lagged Y variables.
	Grow(const cTadd)
	Grows the database sample size.
	Init()
	Resets everything.
virtual	InitData()
	Get the data (first stage prior to estimation).
virtual	InitPar()
	Initializes the parameter values (second stage prior to estimation).
virtual	IsUnivariate()
	Returns TRUE if the model is univariate.
virtual	LoadOptions()
	Used by OxPack to loads the options from persistent storage.
	MapParToFree(const vPar)
	Return the free parameters from the specified full parameter vector.
	Modelbase()
	Constructor.
virtual	Output()
	Prints estimation output.
	OutputHeader(const sTitle)
	Prints the header.
	OutputLogLik()
	Prints the loglikelihood and related statistics.
	OutputMax(const sMethod, const iResult, const vPstart, const bNumerical)
	Prints the maximization settings used.
	OutputPar()
	Prints estimated parameters.
static	PrintTestVal(const dTest, const cR, const cTdf, const sLabel)
	Prints a test statistic and its significance.
virtual	ReceiveData()
	Called by OxPack to let the object receive the data; the object
virtual	ReceiveMenuChoice(const sDialog)
	Called by OxPack to let the object receive the menu choice; the object
virtual	ReceiveModel()
	Called by OxPack to let the object receive the model specification; the object
	ResetFixedPar()
	Reset the values of the fixed parameters.
virtual	SaveOptions()
	Used by OxPack to store the options in persistent storage.
	Select(const theGroup, const aSel)
	Select variables by name into a group.
	SelectByIndex(const theGroup, const iVar, const iLag0, const iLag1)
	Select variables by name into a group.
virtual	SendFunctions()
	Sends functions to dialog (used by DPD, e.g.).
virtual	SendMenu(const sMenu)
	Sends a menu specification to OxPack.
virtual	SendSpecials()
	Sends the names of special variables to OxPack for batch code.
virtual	SendVarStatus()
	Sends the specification of variable status types to OxPack for batch code.
virtual	SetForecasts(const cForc, const bIsLessForecasts)
	Set the number of forecasts.
	SetFreePar(const vParFree)
	Set the free parameters.
virtual	SetMethod(const iMethod)
	Sets the estimation method.
virtual	SetModelSettings(const aValues)
	Called by OxPack to update the estimation settings in the object.
	SetModelStatus(const iModelStatus)
	Sets the model status.
	SetPar(const vPar)
	Sets the parameter values (free and fixed) .
	SetParCount(const cPar, ...)
	Set the number of parameters q (including fixed parameters).
	SetPrint( fPrint)
	Switches printing on or off.
virtual	SetRecursive(const bSet, const cInit)
	Set the number of recursive steps to initiate recursive estimation.
	SetResult(const iResult)
	Sets the result of estimation.
	SetSelDates(const iYear1, const iMonth1, const iDay1, const iYear2, const iMonth2, const iDay2)
	Select a sample by year, month, day.
	SetSelSample(const iYear1, const iPeriod1, const iYear2, const iPeriod2)
	Select a sample (fixed frequency).
	SetSelSampleByDates(const dDate1, const dDate2)
	Select a sample by date values.
virtual	SetStartPar(const vParFree)
	Set the starting values.
	SetVarEx(const mNew, const asNew, const iToffset=0, const viTindex=<>, const iPrint=0)
	Sets variable(s) in the database, used to store estimation results or forecasts.
static	ShowBanner(const bSet)
	Switch off/on printing of startup banner.
virtual	StoreEstimates(const mX, const asX, const iPrint=0)
	Stores estimation results, taking the estimation sample into account.
virtual	StoreForecasts(const mX, const asX, const iPrint=0)
	Stores forecasts, taking the estimation sample into account.
virtual	TestRestrictions( mR, ...)
	Tests a linear restriction.

Modelbase data members (protected by default)
static	sm_bBanner

	m_mCovar

	m_mCovarRobust

	m_iDeterministic

	m_vFixedPar

	m_dForcCrit

	m_vIdxFreePar

	m_vIsFreePar

	m_dLogLik

	m_fMaxManual

	m_iMethod

	m_iModelStatus

static	sm_iOxCode

	m_cPar

	m_vPar

	m_cParFree

	m_fPrint

	m_bRecursive

	m_iResult

	m_cT

	m_iT1est

	m_iT2est

	m_cTforc

	m_viTindex

	m_cTinit

	m_mX

	m_cX

	m_mY

	m_cY

	m_cYlag

Modelbase enum members (protected by default)
	{STATUS_ENDOGENOUS, STATUS_SPECIAL, STATUS_MULTIVARIATE, STATUS_TRANSFORM, STATUS_GROUP, STATUS_ONEONLY, STATUS_GROUP2, STATUS_MULTIPLE, STATUS_DEFAULT, STATUS_GROUP3, STATUS_GROUP4, STATUS_GROUP5}
	OxPack model formulation variable types.
	{SPECIAL_CONSTANT, SPECIAL_TREND, SPECIAL_SEASONAL, SPECIAL_CSEASON}
	OxPack model formulation special variables.
public:	{Y_VAR, X_VAR}
	Types of variables in the model.
public:	{MS_NONE, MS_DATA, MS_PARAMS, MS_ESTIMATED, MS_EST_FAILED}
	Model estimation status.

exported functions

exported functions
	EvalTestRestrictions(const sResTestCode, const cP, const asMsg, const avResMap, const amResJac)

exported enumerations

exported enumerations
	{CTL_LABEL, CTL_GROUP, CTL_SUBGROUP, CTL_CHECK, CTL_INT, CTL_INTRANGE, CTL_RADIO, CTL_RADIOBOX, CTL_SELECT, CTL_DOUBLE, CTL_DATE, CTL_STRING, CTL_EDITOR, CTL_FILE, CTL_FILESAVE, CTL_FOLDER, CTL_MATRIX, CTL_STRMAT, CTL_CHECKLIST, CTL_SAMPLEINDEX, CTL_SAMPLERANGE, CTL_DATEINDEX, CTL_DATERANGE, CTL_ENABLER, CTL_DISABLER, CTL_VARIABLE, CTL_LISTBOX}
	dialog control types.

virtual Modelbase::Batch(const sBatch, ...)

Parameters

sBatch	string, batch function
optional	arguments for the batch function

Returns

0 for error

virtual Modelbase::BatchMethod(const sMethod)

Parameters

sMethod

string

Returns

the current method

virtual Modelbase::Buffering(const bBufferOn)

Parameters

bBufferOn

TRUE or FALSE

virtual Modelbase::ClearEstimation()

Comments: Sets m_mCovar and m_mCovarRobust to -1, m_dLogLik to .NaN and m_iResult to MAX_NOCONV.

virtual Modelbase::ClearModel()

Comments: Sets the model status to and calls ClearEstimation

virtual Modelbase::Covar()

Comments: This function should set m_mCovar, and optionally m_mCovarRobust. It is called when GetCovar GetStdErr (or their robust versions) are called and m_mCovar equals -1 (the default after ClearEstimation). This procedure ensures that the covariance is only computed once when required, and not at all when not required (in some Monte Carlo experiments, for example).

Modelbase::DbDrawTMatrix(const iArea, const mYt, const asY, const iT1, ...)

Parameters

or a string (when only one variable to graph) (optional argument, default is 0) Or vector with value for each row of data. (optional, default is 2) Each subsequent row will have the next index. Or vector with value for each row of data.

iArea	int, area index
mYt	m x T matrix with m y variables
asY	array of strings (holds variable names), or 0 (no names),
mnYear	int, year of first observation (optional argument, default is 1)
iSymbol	int, 0: draw line, 1: draw symbols, 2: draw both
iIndex	int, line index for first row,

Modelbase::DeSelect()

Comments: Also calls ClearModel

Modelbase::DeSelectByIndex(const iSel)

Parameters

iSel	int, selection indices of variable to remove

Comments

Also calls ClearModel

Modelbase::DeSelectByName(const sVar, const theGroup, const iLag)

Parameters

sVar	string, name of variable to remove
theGroup	int or string, group identifier
iLag	int, lag length

Comments

Also calls ClearModel

virtual Modelbase::DoEstimateDlg(const iFirstMethod, const cMethods, const sMethods, const bForcAllowed, const bRecAllowed, const bMaxDlgAllowed)

Parameters

iFirstMethod	int, first possible method index
cMethods	int, number of available methods
cMethods	string with methods, separated by \| (number must match cMethods)
bForcAllowed	TRUE: add field for m_cTforc
bRecAllowed	TRUE: add fields for m_bRecursive and m_cTinit
bMaxDlgAllowed	TRUE: add field for m_fMaxManual

Returns

1 if dialog accepted by user, 0 for cancel or error, -1 to return to formulation

virtual Modelbase::DoEstimation( vParFree)

Parameters

Estimate

vParFree

px1 vector with free parameter values

Returns

px1 vector with parameter values (both fixed and free) or array[3] with extended information: { px1 parameters, string with procedure name, TRUE or FALSE for numerical derivatives }

Comments

This function must be overridden with the actual estimation code if Estimate is used.

virtual Modelbase::DoFormulateDlg(const iLagMode, ...)

Parameters

iLagMode	-1: no lags; >= 0: default for first lag
iLagDefault	(optional) int, value for lag value box.

virtual Modelbase::DoOption(const sOpt, const val)

Parameters

DoOptionsDlg

sOpt	string, option label
val	option value

virtual Modelbase::DoOptionsDlg(const aMoreOptions)

Parameters

aMoreOptions

array[][] with extension of the dialog specification

Returns

TRUE if dialog accepted by user

virtual Modelbase::DoSettingsDlg()

Returns: 1 if dialog accepted by user, 0 for cancel or error, -1 to return to settings,

virtual Modelbase::Estimate()

Returns: TRUE if successful
Comments: Calls InitData and InitPar if not yet called. Then uses InitData to get the free parameters, calls DoEstimation SetFreePar and Output (unless switched off).

virtual Modelbase::FindGroup(const theGroup)

Parameters

GetGroupLabels).

theGroup

int (returned as is), or string (translated in index in

Returns

int, group identifier (0 if not found).

virtual Modelbase::FindMethod(const theMethod)

Parameters

GetGroupLabels).

theMethod

int (returned as is), or string (translated in index in

Returns

int, method identifier (0 if not found).

Modelbase::FixPar(const iP, const dFix)

Parameters

iP	int or vector, index of parameter to fix
dFix	double or vector, value to fix parameter at

Comments

This is used to fix a parameter at the specified value. Subsequently, this parameter is omitted from the vector returned by GetFreePar

Modelbase::ForceYlag(const iYgroup)

Parameters

iYgroup to iYgroup; array[2]: reclassify lagged iYgroup[0] variables to iYgroup[0] but only if they have type iYgroup[1]; array[3]: iYgroup[2]=-1: returns indices of lagged variables that need changing.

iYgroup

scalar: reclassify the lags of all variables that are

Modelbase::FreePar(const iP)

Parameters

iP	int, index of parameter(s) to free, use -1 to free all

virtual Modelbase::GetBatchCode()

Returns: string

virtual Modelbase::GetBatchDatabase()

Returns: string

Modelbase::GetBatchEquation(const sY, const asX)

Parameters

sY	string, variable type
asX	array of strings with specification for this type

Returns

string

virtual Modelbase::GetBatchEstimate()

Returns: string

virtual Modelbase::GetBatchModel()

Returns: string

virtual Modelbase::GetBatchModelSettings()

Returns: string

virtual Modelbase::GetBatchModule(const sModule, const sPackage)

Returns: string

Modelbase::GetCovar()

Returns: pxp covariance matrix (as set by Covar) or <>
Comments: Calls Covar if m_mCovar equals -1 (so not when it is empty).

Modelbase::GetCovarRobust()

Returns: pxp covariance matrix (as set by Covar) or <> if not set
Comments: Calls Covar if m_mCovar equals -1 (so not when it is empty).

Modelbase::GetForecastData(const iGroup, const mnLag, const mxLag, const cTforc, ...)

Parameters

theGroup	int or string, group indicator (e.g. "Y" or PcGive::Y_VAR)
mnLag	int, start lag
mxLag	int, end lag
cTforc	int, number of forecasts (-1: return all)
iT1forc	int, first forecasts observation (optional, default is m_iT2est+1)

Returns

Matrix with the available forecasts data (or an empty matrix if no there is no data).

Comments

When m_viTindex is set, returns the entire series except for observations selected by m_viTindex (cross-section mode).

Modelbase::GetFreePar()

Returns: px1 matrix with free parameters

Modelbase::GetFreeParCount()

Returns: number of free parameters.

Modelbase::GetFreeParNames()

Returns: array[p] with the names of free parameters

virtual Modelbase::GetGroupLabels()

Returns: array of strings.
Comments: By default this returns {"Y", "X"} for Y_VAR, X_VAR, but the derived class should extend this. This function is used to find the group when it is specified as a string instead of an integer, or printed as a string in GetOxModel

Modelbase::GetInEstimationSample(const iT1=-1, const iT2=-1)

Parameters

of a sample range)

iT1	int, first observation index (or -1 for full estimation sample)
iT2	int, last observation index (or -1 for just a sample point, instead

Returns

string

Modelbase::GetInForecastSample(const iT1, const iT2=-1)

Parameters

of a sample range)

iT1	int, first observation index
iT2	int, last observation index (or -1 for just a sample point, instead

Returns

string

Modelbase::GetLogLik()

Returns: double, the loglikelihood value.

virtual Modelbase::GetLongRunInfo()

Returns: empty array

virtual Modelbase::GetLongRunNames()

Returns: empty array

Modelbase::GetMethod()

Returns: the estimation method.

virtual Modelbase::GetMethodLabel()

Returns: string, the text label of the method

virtual Modelbase::GetMethodLabels()

Returns: array of strings.
Comments: By default this returns {"None", "OLS"} for M_NONE, M_OLS, but the derived class should extend this. This function is used to find the group when it is specified as a string instead of an integer, or printed as a string in GetOxModel

virtual Modelbase::GetModelLabel()

Returns: string the label of the model.

virtual Modelbase::GetModelSettings()

Returns: array[] with {value, label} pairs.

Modelbase::GetModelStatus()

Returns: one of: MS_NONE, MS_DATA, MS_PARAMS, MS_ESTIMATED, MS_EST_FAILED

virtual Modelbase::GetOxCode()

Returns: string with Ox code

virtual Modelbase::GetOxDatabase(const sObj, const sClass)

Parameters

sObj	string, object name
sClass	string, class name

Returns

string with Ox code

virtual Modelbase::GetOxDecl(const sObj, const sClass)

Parameters

sObj	string, object name
sClass	string, class name (GetOxCode uses classname(this) by default)

Returns

string with Ox code

virtual Modelbase::GetOxEstimate(const sObj, const sClass)

Parameters

sObj	string, object name
sClass	string, class name

Returns

string with Ox code

virtual Modelbase::GetOxModel(const sObj, const sClass)

Parameters

sObj	string, object name
sClass	string, class name

Returns

string with Ox code

virtual Modelbase::GetOxModelSettings(const sObj, const sClass)

Parameters

sObj	string, object name
sClass	string, class name

Returns

string with Ox code

virtual Modelbase::GetPackageName()

Returns: string, package name

virtual Modelbase::GetPackageVersion()

Returns: string, package version

Modelbase::GetPar()

Returns: qx1 vector of all parameters

Modelbase::GetParCount()

Returns: the total number of parameters (free and fixed).

virtual Modelbase::GetParNames()

Returns: array[q] with names of all parameters.

Modelbase::GetParStatus()

Returns: array[3] { total number of parameters q, q x 1 matrix with 1 in position of free parameters, and 0 for fixed, q x 1 matrix with fixed value in position of fixed parameters (free positions are unused), }

virtual Modelbase::GetParTypes()

Returns: empty array.

Modelbase::GetPrint()

Returns

virtual Modelbase::GetResVar()

Returns: the nxn matrix with the residual variance (<> in Modelbase) (n equals 1 for univariate models).

virtual Modelbase::GetResiduals()

Returns: TxK matrix with residuals (<> in Modelbase).

Modelbase::GetResult()

Returns: The estimation result.

Modelbase::GetStdErr()

Returns: qx1 vector with standard errors of all parameters (fixed parameters set to 0) or <>.
Comments: Calls Covar if m_mCovar equals -1 (so not when it is empty).

Modelbase::GetStdErrRobust()

Returns: qx1 vector with standard errors of all parameters (fixed parameters set to 0); returns <> if the robust covariance matrix is not available.
Comments: Calls Covar if m_mCovar equals -1 (so not when it is empty).

Modelbase::GetX()

Returns: the TxcX matrix X, m_mX.

Modelbase::GetY()

Returns: the Txn matrix Y, m_mY.

virtual Modelbase::GetcDfLoss()

Returns: degrees of freedom lost (for tailt, tailchi etc.)

virtual Modelbase::GetcT()

Returns: int, the actual number of observations

Modelbase::GetcX()

Returns: number of X variables

Modelbase::GetcY()

Returns: number of Y variables

Modelbase::GetcYlag()

Returns: number of columns in m_mX that are lagged Y variables
Comments: Lagged Y's are at the start of m_mX.

Modelbase::Grow(const cTadd)

Parameters

in: int, number of observations to grow database sample by (> 0: cTadd observations are added at the end; < 0: -cTadd observations are added at the beginning).

cTadd

Returns

vector with database indices of variables for which new values were created (so not those that were just extended with .NaN).

Comments

Newly created observations are set to .NaN, except for variables named Constant, Trend, Seasonal, CSeasonal which are updated (NB. Season* and CSeason* are not extended); moves the estimation sample.

Modelbase::Init()

Resets everything.

virtual Modelbase::InitData()

Returns: TRUE if successful
Comments: If successful sets model status to MS_DATA.

virtual Modelbase::InitPar()

Returns: TRUE if successful
Comments: If model status is less than MS_DATA InitData is called. If successful sets model status to MS_PARAMS.

virtual Modelbase::IsUnivariate()

Returns: TRUE if the model is univariate

virtual Modelbase::LoadOptions()

Used by OxPack to loads the options from persistent storage.

Modelbase::MapParToFree(const vPar)

Parameters

vPar	q vector with parameter values (both fixed and free)

Returns

p x 1 vector with free parameter values.

Modelbase::Modelbase()

Constructor.

virtual Modelbase::Output()

Comments: Calls OutputHeader, OutputPar, OutputLogLik

Modelbase::OutputHeader(const sTitle)

Parameters

sTitle

string title of header

Modelbase::OutputLogLik()

Prints the loglikelihood and related statistics.

Modelbase::OutputMax(const sMethod, const iResult, const vPstart, const bNumerical)

Parameters

sMethod	string, maximization method
iResult	int, maximization result
vPstart	vector with starting values (or <> to skip)
bNumerical	int, {\tt TRUE} if using numerical derivatives

Modelbase::OutputPar()

Prints estimated parameters.

static Modelbase::PrintTestVal(const dTest, const cR, const cTdf, const sLabel)

Parameters

dTest	test statistic
cR	first degrees of freedom
cTdf	second degrees of freedom (0 for chi^2, otherwise F-distribution)
sLabel	name of test\\

virtual Modelbase::ReceiveData()

Called by OxPack to let the object receive the data; the object will ask for the data in this function and store it in the underlying database.

virtual Modelbase::ReceiveMenuChoice(const sDialog)

Parameters

sMenu

string, menu identifier label

Returns

1 if successful

virtual Modelbase::ReceiveModel()

Called by OxPack to let the object receive the model specification; the object will ask for the model in this function and store the selection in the underlying database.

Modelbase::ResetFixedPar()

Reset the values of the fixed parameters.

virtual Modelbase::SaveOptions()

Used by OxPack to store the options in persistent storage.

Modelbase::Select(const theGroup, const aSel)

Parameters

followed by start_lag, end_lag (e.g. {"CONS",0,3, "Constant", "INC",0,1}

theGroup	int or string, group identifier (e.g. "Y" or PcGive::Y_VAR)
aSel	array of variable names, each optionally

Comments

Invalidates the selection sample. Also calls ClearModel

Modelbase::SelectByIndex(const theGroup, const iVar, const iLag0, const iLag1)

Parameters

theGroup	int or string, group identifier (e.g. "Y" or PcGive::Y_VAR)
iVar	int or vector[k] column index or indices
iLag0	common start lag or vector[k] with starting lag lengths
iLag1	common end lag or vector[k] with ending lag lengths

Comments

Invalidates the selection sample. Also calls ClearModel

virtual Modelbase::SendFunctions()

Sends functions to dialog (used by DPD, e.g.).

virtual Modelbase::SendMenu(const sMenu)

Parameters

sMenu

string: "Model" or "Test" menu.

virtual Modelbase::SendSpecials()

Comments: Also used by DoFormulateDlg

virtual Modelbase::SendVarStatus()

Comments: Also used by DoFormulateDlg

virtual Modelbase::SetForecasts(const cForc, const bIsLessForecasts)

Parameters

the estimation sample is unaffected.

cForc	sets the number of forecasts
bIsLessForecasts	TRUE: the estimation sample is reduced by the number of forecasts, FALSE:

Modelbase::SetFreePar(const vParFree)

Parameters

vParFree

p x 1 vector with free parameter values.

virtual Modelbase::SetMethod(const iMethod)

Parameters

theMethod

int or string with estimation method

virtual Modelbase::SetModelSettings(const aValues)

Parameters

aValues

array[] with {value, label} pairs.

Modelbase::SetModelStatus(const iModelStatus)

Parameters

iModelStatus,

one of: MS_NONE, MS_DATA, MS_PARAMS, MS_ESTIMATED, MS_EST_FAILED

Modelbase::SetPar(const vPar)

Parameters

vPar	qx1 vector with parameter values (both fixed and free)

Comments

Also sets the parameter count if this is still zero.

Modelbase::SetParCount(const cPar, ...)

Parameters

cPar	int, parameter count
bAdd	(optional) TRUE: add parameters to current count; else set the count.

Comments

Must be called as part of the estimation procedure.

Modelbase::SetPrint( fPrint)

Parameters

fPrint

TRUE or FALSE

virtual Modelbase::SetRecursive(const bSet, const cInit)

Parameters

bSet	True or FALSE
cInit	int, initial observations for recursive estimation

Modelbase::SetResult(const iResult)

Parameters

iResult

int, usually result of MaxBFGS

Modelbase::SetSelDates(const iYear1, const iMonth1, const iDay1, const iYear2, const iMonth2, const iDay2)

Parameters

iYear1	int, start year of selection
iMonth1	int, start month of selection
iDay1	int, start day of selection
iYear2	int, end year of selection
iMonth2	int, end month of selection
iDay2	int, end day of selection

Returns

Number of observations in the sample.

Comments

Also calls ClearModel

Modelbase::SetSelSample(const iYear1, const iPeriod1, const iYear2, const iPeriod2)

Parameters

iYear1	int, start year of selection, use -1 for earliest year and period
iPeriod1	int, start period of selection
iYear2	int, end year of selection, use -1 for latest year and period
iPeriod2	int, end period of selection

Returns

Number of observations in the sample.

Comments

Also calls ClearModel

Modelbase::SetSelSampleByDates(const dDate1, const dDate2)

Parameters

dDate1	date value, start date
dDate2	date value, end date

Returns

Number of observations in the sample.

Comments

Also calls ClearModel

virtual Modelbase::SetStartPar(const vParFree)

Returns: TRUE if successful
Comments: CallsInitPar, then sets the free parameters as specified.

Modelbase::SetVarEx(const mNew, const asNew, const iToffset=0, const viTindex=<>, const iPrint=0)

Parameters

mNew	matrix[][k]
asNew	string or array[k] of strings.
iToffset	(optional) first index in database to store to (default is 0).
viTindex	(optional) database observation indices to store to (default is <>).
iPrint	(optional) prints information about the action (default is 0).

Comments

If the variable already exists, then observations outside the sample are not changed (unlike SetVar). If necessary, the database is extended to accomodate the observations (like SetVar).

static Modelbase::ShowBanner(const bSet)

Switch off/on printing of startup banner.

virtual Modelbase::StoreEstimates(const mX, const asX, const iPrint=0)

Parameters

mX	matrix[T][k] with estimation results (e.g. residuals)
asX	array[k] with names
iPrint	prints storage report (default 0)

Comments

Results are stored from m_iT1est onwards, unless m_viTindex is set, in which case they are stored for all observations in the estimation sample. Uses SetVarEx

virtual Modelbase::StoreForecasts(const mX, const asX, const iPrint=0)

Parameters

mX	matrix[T][k] with forecasts
asX	array[k] with names
iPrint	prints storage report (default 0)

Comments

Forecasts are stored from m_iT2est + 1 onwards (i.e. immediately following the estimation sample), unless m_viTindex is set, in which case they are stored for all observations <em>not</em> in the estimation sample. Uses SetVarEx

virtual Modelbase::TestRestrictions( mR, ...)

Parameters

of free parameters that are set to zero in the test, or a vector of free parameter indices. linear restrictions and vR an sx1 or 1xs vector with values to which the restrictions should add. Unless vR==1 in which case the tests are on the longrun.

mR	with this the only argument: an array of strings with names
vR	(optional) if this argument is given, mR is an sxp matrix with
sGenRes	(optional), string with general restrictions

Returns

Comments

GetFreeParNames, GetFreePar GetCovar

EvalTestRestrictions(const sResTestCode, const cP, const asMsg, const avResMap, const amResJac)