3. Exploratory data analysis methods

This chapter emphasizes the need to carry out Exploratory Data Analysis (EDA) before embarking on any modeling endeavor. EDA techniques introduced include non-spatial techniques such as histogram, pairwise scatter plots; spatial methods such as variogram and Kriging; and temporal exploration such as time series plots. In particular, the chapter explores the following four data sets which are provided in the bmstdr package. nyspatial : point referenced spatial air pollution data from New York, nysptime: point referenced spatio-temporal air pollution data from New York, engtotals: areal unit data on Covid-19 mortality in England, engdeaths: areal unit weekly temporal data on Covid-19 mortality in England, Familiarity of these data sets is crucial for understanding the modeling performed in the later chapters.

R code for reproducing the figures in Chapter 3.

england_jsa_dec_2019.pngboxplots_sptime.pngcovd_and_socio.pngendeaths_socio_eco.pngeng_boxplot_case_rate.pngengland_case_rate.pngengland_case_smr.pngengland_house_price_mar_2020.pnghist_nyspatial.pngkriged_ny.pnglog_pop_density.pngmean_no2_by_LA.pngmean_variance_nysptime.pngno2_boxplots.pngnysptime_edaplots.pngpairs_nyspatial.pngsmr_covid_death.pngsptDyn_dynamic_effects.pngsptDyn_spat_effects.pngsqrto3_v_maxtemp_nysptime.pngsqrto3_v_rh_nysptime.pngsqrto3_v_wdsp_nysptime.pngtarget_importance.pngtimeseriesplot_nysptime.pngvalidation_time_series.pngvariogram_nyspatial.png