|
|
Using the Estimation Types screen of the Estimate dialog. |
ESTIMATE: Estimation
To access this dialog:
-
-
Type 'ESTIMATE' in the Command window. In the Estimate dialog, select Next until you reach the Estimation Types tab.
The Estimation Types screen of the Estimate dialog is used to define the method by which a grade estimation will be performed. This information will be stored in the Grade Estimation Parameter File.
It is possible to select different grades to estimate, using different methods and different parameters all in a single run of Estimate. The different combinations of grades/methods/parameters etc. are controlled by adding estimation items to the Index list on the left, as required, and then editing the default values for each item.
This right-hand side of the dialog comprises the following sub-tabs containing collections of fields relating to a particular aspect of search volume definition.
-
Attributes: set up the general parameters for each estimation. More...
-
Options: select the anisotropy method to be used during estimation. More...
-
Indicator Estimation: select the mean grade method, output files, order relation correction and cutoff parameters if indicator estimation is to be performed. More...
-
Summary: shows a summary of the Estimation Parameters file. More...
The Estimation Types list contains the following controls:
Estimation Types list: select existing estimations to edit or delete. Selection of an item updates the relevant fields on the tabs on the right hand side. The right-click menu gives a range of options:
-
Add: add a new estimation at the bottom of the list. The default estimation description for estimation number n is “Estima Param N”. It is recommended that you edit the description to make it meaningful for your data. You can do this by highlighting the estimation and selecting Edit Description from the right-click menu
-
Copy: copy the estimation parameters into the buffer, so they can be pasted.
-
Cut: copy the estimation parameters into the buffer, so they can be pasted, and delete the selected estimation.
-
Paste: replace the parameters of the selected estimation with the values in the Copy/Cut buffer.
-
Delete: delete the selected estimation.
-
Edit Description: edit the description of the selected estimation.
-
Reset all Estimation Parameters: erase all estimations, after confirmation.
-
Up: move the selected estimation up one place.
-
Down: move the selected estimation down one place.
-
Sort by Reference Number: reorder the current list. Note that the field used to define the reference number is contained in brackets after the menu item.
-
Annotate All Using Description: only the description will be displayed, not the number. Note that the field used to define the description is contained in brackets after the menu item.
Add button: create a new search volume with default values, and add it to the end of the current list. This option is also available from the right-click menu.
Delete button: delete the currently selected search volume item from memory. This option is also available from the right-click menu.
Reset button: erase all search volumes, after confirmation. This option is also available from the right-click menu.
Import button: the Import button allows you to import estimation parameters from an existing Estimation Parameter file. All sets of estimation parameters from the file will be imported. The imported estimations will retain their numbers unless they clash with an existing number. In that case the estimation being imported will be renumbered as the maximum existing estimation number plus 1. Note that estimation numbers are integers.
Export button: the Export button allows you to save the current set of estimation parameters in an Estimation Parameter file. All parameters will be exported.
The Attributes sub-tab is used to define the general parameters for the selected estimation:
Estimation Method (Nearest Neighbour, Inverse Power....): select a radio button to define the general estimation method. The available estimation types are:
-
Nearest Neighbour: select this option to perform a Nearest Neighbour grade estimation.
-
Inverse Power of Distance: select this option to perform an IPD grade estimation. The Dynamic Anisotropy option requires angle fields to be interpolated into the block model. This is selected using the Angle Data check box.
-
Ordinary Kriging: select this option to perform an Ordinary Kriging grade estimation. You have the additional option of selecting log kriging.
-
Simple Kriging: select this option to perform an Simple Kriging grade estimation. You have the additional option of selecting log kriging.
-
Sichel's T Estimator: select this option to use Sichel's T Estimator method.
-
F-value: use the average value of the variogram in the cell.
-
Lagrange multiplier: use the Lagrange multiplier calculated when solving the Ordinary Kriging matrix.
(Select a link above for more information on each method, or for general information about the available methods, see Grade Estimation Methods).
Indicator Estimation: if Indicator Estimation (IE) is selected, then it applies to all estimates.
-
You cannot mix IE with any non-indicator methods
-
For IE, you must define an estimate in the Index panel for each cutoff.
-
If zone control is being performed, you must define an estimate for each cutoff and each zone value. The cutoffs may vary between zones.
All the indicator parameters are defined on the Indicator Estimation sub-tab, which is greyed-out unless Indicator Estimation is selected here. Selecting this option ensures that the INDEST process will be accessed during estimation.
Data Fields: select the sample and model grade field files (or select the Same as Sample check box to define one file name for both purposes). For more information on Data Fields, see VALUE_IN and VALUE_OU.
Model Fields: define the secondary fields required (Number of Samples, Variance, Transformed Distance, Search Volume). For more information on secondary fields, see Grade Estimation Secondary Fields.
Search and Variogram Definition: each drop-down list in this section contains all currently defined search volumes and variogram model definitions (defined on the respective tab of the Estimate dialog). Depending on your selection of estimation method (see above), you may not have access to the Variogram Model drop-down list. For more information on these parameters, see Grade Estimation Parameter File.
Zone Field Values: if one or two zone control fields have been specified on the Files tab, a different set of estimation parameters can be assigned to each zone value (one zone control field), or combination of zone values (two zone control fields). For more information on zonal control, see Grade Estimation Zonal Control.
The Options sub-tab is used to define the additional parameters for the selected estimation. The fields shown will depend on the Estimation Method selected on the Attributes tab (see above), however, the list below represents an exhaustive tally of all possible fields for all methods, with an indication of the methods each field is associated with.
Anisotropy method: select either Search Ellipse, Isotropic or User Defined options. For more information on anisotropy methods, see Grade Estimation Nearest Neighbour.
Available for methods: Nearest Neighbour, Inverse Power of Distance.
Distance Weighting: select distance weighting options (for more information on these fields, see Grade Estimation Inverse Power of Distance:
-
Include offset distance: select this check box to specify a Distance (see below).
-
Distance: if a sample lies exactly on a discretised point, then it will be at zero distance from that point and will get 100% of the weight. This can lead to a biased estimate, particularly if there is only one discretised point and there are other samples lying within the cell. However, this can be avoided by specifying a positive value for the field Distance.
-
Power: select a value to define the power of distance for IPD weighting.
Distance Weighting fields are available for methods: Inverse Power of Distance.
Weighting Fields: define the length and density weighting fields using the supplied drop-down lists. For more information, refer to IPD Length and Density Weighting.
Weighting fields are available for methods: Inverse Power of Distance.
Negative Kriging Weights: select the Reset Negative Weights to zero with this option. For more information on handling negative weights, see Grade Estimation Negative Kriging Weights.
Available for methods: Ordinary Kriging (with or without log option), Simple Kriging (with or without log option)
Local Mean Value: for Simple Kriging, you can elect to either calculate the local mean value as the arithmetic mean of all samples lying in the search volume (Calculate Mean), or provide a field description from which it can be derived from the model (Use Model Field, and select a Local Mean Field).
Available for methods: Simple Kriging (with or without log option)
Log Kriging Settings: select your log kriging option from either Rendu's Approximation or General Case. For more information on log kriging, refer to Grade Estimation Lognormal Kriging.
Available for methods: Ordinary Kriging (with log option), Simple Kriging (with log option)
Additive Constant: specify an additive constant for Sichel's T Estimator method. This will be added to the Grade Estimation Parameter file as the ADDCON field. You should be aware, however, that this field is also used by the IPD method, but in a different context.
Available for methods: Sichel's T Estimator
Indicator Estimation Tab Field Details:
The Indicator Estimation sub-tab will be disabled unless Indicator Estimation was selected on the Attributes sub-tab (see above).
Mean Grade Method: the calculation of the grade above each cutoff requires that the average grade between each successive pair of cutoffs be specified. The following options are available:
-
Average of lower and upper cutoffs: average of minimum and maximum cutoff values. The grade above the highest cutoff is calculated as the highest cutoff plus half the difference between the highest and second highest cutoffs.
-
Average from sample file for all cutoff intervals: select this option to ensure that the mean grade is calculated by the INDEST process from the grades of the samples in the IN file, for all cutoff intervals.
-
Average from sample file; median value for samples above top cutoff: select this option to use a median average above the top cutoff value.
-
Manual Entry: values are specified by the user, using the BINGRADE and ABVGRADE fields in the ESTPARM file. The BINGRADE field contains the grade below the cutoff and the ABVGRADE field the grade above the cutoff. The ABVGRADE field is therefore only used for the top bin, and can be entered into the Manual Entry: Average grade above top cutoff field (see below).
|
Although the option appears on the Indicator Estimation tab for every estimate, your selection applies to all estimates. |
Output Files: two optional output files may be defined:
-
Average grade for each cutoff interval: this output file contains cutoff grade ranges and average grade used for each range. It will include zone field(s), if any, plus the following fields:
BIN: bin or grade range number
LO_CUT: lower cutoff grade
UP_CUT: upper cutoff grade
NSAMPLES: number of samples in IN file lying within the bin
BINGRADE: bin grade used for indicator estimation. This is dependent on the Mean Grade Method .
SAMPMEAN: mean grade of samples in IN file lying within the bin -
Indicator values for each cutoff: the output indicator file is a copy of the sample input IN file, but also includes the 0/1 indicator values for each cutoff.
Include "Probability Above" and "Grade Above"...: select this check to ensure these fields are included in the output Grade model.
Order Relation Correction: one of the main drawbacks of the indicator estimation method is the "Order Relation" Problem. This will occur if the proportion of the sub-cell above cutoff n is estimated to be less than the proportion above cutoff n+1. e.g., if PRAB(n) < PRAB(n+1). Three options are available to correct for this:
-
Downwards
-
Upwards
-
Average of Downwards and Upwards
The recommended method (and default) is method 3.
Cutoff Data: the following options exist for defining the cutoff (indicator) grade:
-
Upper cutoff grade for current interval: the cutoff (or indicator) grade must be specified for each estimation.
-
Mean grade between current and previous cutoffs: you must also define the mean grade between the current and previous cutoff for each estimation. Although, this option appears for every estimate, your selection applies to all estimates.
-
Manual Entry: Average grade above top cutoff: if Manual Entry has been selected for the Mean Grade Method (see above), this option allows you to define the mean grade for the top bin manually.
The Summary tab shows a tabular view of the data columns and values found within the current Estimation Parameter File. For more details on these fields, see Grade Estimation Parameter File.
|
|
The following buttons are also available along the bottom of this screen: Restore: restores the settings added prior to the lastRunof theEstimatedialog. This will restore all settings on all screens. Clear: reset all fields on all screens to their default values. Previous/Next: move backwards or forwards one screen in the order of Estimate screens Run: run the current grade estimation, launching theESTIMAprocess. Cancel: cancel theESTIMATEprocess without running an estimation. |
