Some Robust Statistical Procedures and Their Application to Air Pollution Data

Application of Some Robust Statistical Procedures to Air Pollution Data in the San Francisco Bay Area

“…Images are produced from the best available original document. In a paper by Cleveland and Guarino [1], the authors investigated production and origin of ozone within the New York City-northern New Jersey region. Biweight regression analysis was used to relate daily maximum ozone concentrations at pairs of sites, and the 10% trimmed means was used to examine the diurnal variability of concentration.…”

“…The equivalent method used by Cleveland and Guarino[1] is similar to the above procedure except in the iteration procedure < -1 -1.=f jvalue of the residual on k-th K,n day in the n-th iteration step. 1. r' = median of {rk3 The fitted line from BIWEIGHT (biweight regression fitting) after the fifth iteration is LOG(Y) = -1.82 + 0.077 31 + 0.044 x.…”