The treatment of missing ‘not at random’ geological data for ore grade modelling

Abstract

Geostatistics aims to spatially model multiple correlated attributes with some under-sampled variables. Missing samples can distort the data set statistics, and for the model to be consistent, it is necessary to understand the mechanisms that cause the missing data (MD). There are three mechanisms that can govern MD: Missing Completely at Random (MCAR), Missing at Random (MAR) and Missing Not At Random (MNAR). More specifically, when there is a systematic difference between measured and the missing samples, the data is subject to the MNAR mechanism, which leads to biases in the final model if not dealt with adequately. A case is presented where an under-sampled attribute, U (ppm), is modelled to infer its values over the study area through co-simulation approach and an univariate simulation performed on an artificially complete set, considering the MNAR mechanism, which shows a reduction in the relative estimation errors compared to the co-simulated model.

Keywords

MNAR data imputation grade modelling simulation Co-simulation

Get full access to this article

View all access options for this article.

References

Barnett

, Deutsch

2008. Imputation of geologic data (Annual Report). Edmonton (AL): University of Alberta.

Barnett

2011. Tools for multivariate geostatistical modeling. Guidebook Series. Vol. 13. Edmonton (AL): Centre for Computational Geostatistics (CCG).

Barnett

, Deutsch

2012. Missing data replacement in a complex multivariate context (Annual Report). Edmonton (AL): University of Alberta.

Barnett

, Manchuck

, Deutsch

CV.

2014. Projection pursuit multivariate transform. Math Geosci. 46(3):337–359. doi: 10.1007/s11004-013-9497-7

Chilès

, Delfiner

1999. Geostatistics: modeling spatial uncertainty. New York (NY): Wiley.

Cohen

1988. Statistical Power analysis for behavioral sciences. 2nd ed. New York (NY): Erlbaum Publ.

Deutsch

, Journel

1992. GSLIB: a geostatistical software library and user's guide. New York (NY): Oxford University Press.

Diggle

, Menezes

, Su

2010. Geostatistical inference under preferential sampling. J R Stat Soc. 59(2):191–232. doi: 10.1111/j.1467-9876.2009.00701.x

Doyen

, Den Boer

, Pillet

1996. Seismic porosity mapping in the ekofisk field using a new form of collocated crokriging. Proceedings of the SPE Annual Technical Conference and Exhibition; Oct 6–9; Denver, CO. Society of Petroleum Engineers. SPE paper number 36498. p. 21–30.

10.

Enders

2010. Applied missing data analysis. New York (NY): The Guilford Press.

11.

Goovaerts

1997. Geostatistics for natural resources evaluation. New York (NY): Oxford University Press.

12.

Isaaks

, Srivastava

1989a. An introduction to applied geostatistics. New York (NY): Oxford University Press.

13.

Isaaks

, Srivastava

1989b. Data from: An Introduction to Applied Geostatistics [Walker Lake data set]. 52north.org Digital Repository. [accessed 2017 Jun 28]. wiki.52north.org/AI_GEOSTATS/AI_GEOSTATSData.

14.

Isaaks

1990. The application of Monte Carlo methods to the analysis of spatially correlated data (PhD thesis). California (CA): Stanford University.

15.

Little

, Rubin

2002. Statistical analysis with missing data. New York (NY): Wiley.

16.

Marechal

1970. Cokriegeage et regression en correlation intrisique. Fontainebleau: Centre de geostatistique.

17.

Matheron

1963. Principles of geostatistics. Econ Geol. 58(8):1246–1266. doi: 10.2113/gsecongeo.58.8.1246

18.

Pyrcz

, Deutsch

2002. Geoestatistical reservoir modeling. New York (NY): Oxford University Press.

19.

Ren

2007. Exact downscaling in reservoir modeling (PhD thesis). Edmonton (AL): University of Alberta.

20.

Rosenblatt

1952. Remarks on multivariate transformation. Ann Math Stat. 23(3):470–472. doi: 10.1214/aoms/1177729394

21.

Rossi

, Deutsch

2014. Mineral resource estimation. New York (NY): Springer.

22.

Rubin

DB.

1976. Inference of missing data. Biometrika. 63(3):51–59. doi: 10.1093/biomet/63.3.581

23.

Rubin

1987. Multiple imputation for response in surveys. New York (NY): Wiley.

24.

Rubin

1988. An overview of multiple imputation. Proceedings of the Survey Research Section. New Orleans (LA): American Statistical Association.

25.

Schafer

, Graham

JW.

2002. Missing data: our view of the state of the art. Psychol Methods. 7(2):147–177. doi: 10.1037/1082-989X.7.2.147

26.

da Silva

, Costa

JFCL

2016. Missing data: How to proceed. Paper presented at the 10th International Geostatistical Congress; Sep 5–9; Valencia, Spain.

27.

Verly

1993. Sequential gaussian co-simulation: a simulations method integrating several types of information. In: Sores A, editor. Qantitative Geology and Geostatistics, Berlin: Springer-Verlag, p. 543–554.