Beyond Where: Modeling Spatial Relationships and Making Predictions

Подписаться 51 тыс.

Просмотров 12 тыс.

50% 1

Once we’ve identified where patterns are present, the next logical question is “why?” This workshop will cover techniques for examining, modeling and exploring our spatial data to uncover relationships and predict spatial outcomes. Application and use of generalized linear regression (GLR) and geographically weighted regression (GWR) will be demonstrated, including a look at new functionality to model both count and binary data using poisson and logistic models. You will learn how to build a model and how to effectively interpret the results and diagnostics
--------------------------------------------------------------------------------------------------------------------------
Follow us on Social Media!
Twitter: / esri
Facebook: / esrigis
LinkedIn: / esri
Instagram: / esrigram
The Science of Where: www.esri.com

Наука

Опубликовано:

7 июл 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 5

@TheTunamouse 3 года назад

This is incredibly useful, really clearly and simply explained - thank you very much!

@GoogMok Год назад

Good presentation ! where can we found the datasets used here ! Thanks !

@ecare-s9160 Год назад

Hello, please I am working on spatial data. I am trying to map Soil organic carbon using random forest and MLR . I have my Soil organic data and covariates. How do I merge them to be in the same grid format for my RF and MLR?

@chelseadelgrande 2 года назад

The documentation says there should be charts with the binary GLR, but they aren't appearing for me and they don't appear in this demo. Also the GLR diagnostic are much less comprehensive for binary than shown in this example. I only get AICC in the messages. Would love to know how you're supposed to test the model accuracy without the information you need.

@chelseadelgrande 2 года назад

For anyone coming here for help, there are two separate GLR tools that provide separate outputs. GLR Spatial Analyst and GLR Geoanalyst....super confusing, but GLR Spatial Analyst provides the charts and model strength info, and GLR Geoanalyst seems to run faster and just gives the coefficient table.