Re: MI Hotspot Thresholds
Good stuff for spatial analysis in MapInfo Pro. You might also want to consider SoilRx as well. Does nifty 3-D, variograms, surface generation, units conversion, and more... There is a tutorial, system download, and demo data at the site. FYI http://www.redhensystems.com/soilrx MidNight Mapper aka Neil 7/16/99 "W. Kent Treichel" wrote: > I recently completed a Geo Statistics class, so I am certainly no expert, > but the problem sounds familiar. We used a variety of software, most of it > free. You can use Variowin to compute the variogram which provides > distance and direction of correlation between values. Next you can put the > data into GEO Eas, a free DOS app, that will figure the Kriging Standard > Deviation for each grid. That data can then be put into Surfer, free > downloads available, which can map the interpolated values, as well as the > associated probability. It would be in your best interest to verify this > methodology with a third party. > > S.J.Reid wrote: > > > I'm looking for any scripts/extensions/software/techniques > > which can be used on/with Mapinfo/VM to aid in the creating of > > statistically robust thresholds for density/interpolated grids. > > What I'm researching more precisely is crime data, and how to apply > > thresholds for the defining of a crime hotspot. > > The original data is point data, which I create grids from, using > > the different techniques available in Mapinfo/VM. The question comes > > then to how applicable the different values are to setting a > > statistically threshold, and how to do this. > > Good rules of thumb would be welcome, but also autocorrelation > > statistics, or LISA statistics, such as Gi and Gi*, or Morans's i. > > Help would be most appreciated, as I am seeking to research how > > well Mapinfo/VM and ArcView Spatial Analyst are suited to perform > > this task, the results being advice to police cartographers on > > possible steps to follow. > > > >Svein Reid, Msc GIS, University of Edinburgh. > > -- > > To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put > > "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED] > > -- > W. Kent Treichel > Wilder Research Center > 1295 Bandana Blvd N. Suite 210 > St. Paul, MN 55108 > (651) 647-4636 > __ > > We sleep safely in our beds because > rough men stand ready in the night > to visit violence on those > who would do us harm. > > George Orwell > > -- > To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put > "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED] -- To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]
Re: MI Hotspot Thresholds
I recently completed a Geo Statistics class, so I am certainly no expert, but the problem sounds familiar. We used a variety of software, most of it free. You can use Variowin to compute the variogram which provides distance and direction of correlation between values. Next you can put the data into GEO Eas, a free DOS app, that will figure the Kriging Standard Deviation for each grid. That data can then be put into Surfer, free downloads available, which can map the interpolated values, as well as the associated probability. It would be in your best interest to verify this methodology with a third party. S.J.Reid wrote: > I'm looking for any scripts/extensions/software/techniques > which can be used on/with Mapinfo/VM to aid in the creating of > statistically robust thresholds for density/interpolated grids. > What I'm researching more precisely is crime data, and how to apply > thresholds for the defining of a crime hotspot. > The original data is point data, which I create grids from, using > the different techniques available in Mapinfo/VM. The question comes > then to how applicable the different values are to setting a > statistically threshold, and how to do this. > Good rules of thumb would be welcome, but also autocorrelation > statistics, or LISA statistics, such as Gi and Gi*, or Morans's i. > Help would be most appreciated, as I am seeking to research how > well Mapinfo/VM and ArcView Spatial Analyst are suited to perform > this task, the results being advice to police cartographers on > possible steps to follow. > >Svein Reid, Msc GIS, University of Edinburgh. > -- > To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put > "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED] -- W. Kent Treichel Wilder Research Center 1295 Bandana Blvd N. Suite 210 St. Paul, MN 55108 (651) 647-4636 __ We sleep safely in our beds because rough men stand ready in the night to visit violence on those who would do us harm. George Orwell -- To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]
Re: MI Hotspot Thresholds
Svein I am working on similar problems using vertical mapper and trying to develop some rules. I am also comparing the results that I get with VM to STAC hot spot areas. You might also look at SPAM (available at http://www.geog.nottingham.ac.uk/~jerry/mystuff/spam.htm also crimestat being developed for CMRC (crime mapping research center) by Ned Levine calculates statistics such as Moran's I for large crime data sets. Are you a member of the crimemap list serve? ([EMAIL PROTECTED]) Are there other mappers working on similar problems out there? Let's keep in touch. Dick <<< "S.J.Reid" <[EMAIL PROTECTED]> 7/15 7:46a >>> I'm looking for any scripts/extensions/software/techniques which can be used on/with Mapinfo/VM to aid in the creating of statistically robust thresholds for density/interpolated grids. What I'm researching more precisely is crime data, and how to apply thresholds for the defining of a crime hotspot. The original data is point data, which I create grids from, using the different techniques available in Mapinfo/VM. The question comes then to how applicable the different values are to setting a statistically threshold, and how to do this. Good rules of thumb would be welcome, but also autocorrelation statistics, or LISA statistics, such as Gi and Gi*, or Morans's i. Help would be most appreciated, as I am seeking to research how well Mapinfo/VM and ArcView Spatial Analyst are suited to perform this task, the results being advice to police cartographers on possible steps to follow. Svein Reid, Msc GIS, University of Edinburgh. -- To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED] -- To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]
MI Hotspot Thresholds
I'm looking for any scripts/extensions/software/techniques which can be used on/with Mapinfo/VM to aid in the creating of statistically robust thresholds for density/interpolated grids. What I'm researching more precisely is crime data, and how to apply thresholds for the defining of a crime hotspot. The original data is point data, which I create grids from, using the different techniques available in Mapinfo/VM. The question comes then to how applicable the different values are to setting a statistically threshold, and how to do this. Good rules of thumb would be welcome, but also autocorrelation statistics, or LISA statistics, such as Gi and Gi*, or Morans's i. Help would be most appreciated, as I am seeking to research how well Mapinfo/VM and ArcView Spatial Analyst are suited to perform this task, the results being advice to police cartographers on possible steps to follow. Svein Reid, Msc GIS, University of Edinburgh. -- To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put "unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]
