Hydroxyl (OH) and peroxy radicals (HO2 and RO2) were measured in the Pearl River Delta, which is one of the most polluted areas in China, in autumn 2014. The radical observations were complemented by measurements of OH reactivity (inverse OH lifetime) and a comprehensive set of trace gases including carbon monoxide (CO), nitrogen oxides (NOx=NO, NO2) and volatile organic compounds (VOCs). OH reactivity was in the range from 15 to 80 s−1, of which about 50 % was unexplained by the measured OH reactants. In the 3 weeks of the campaign, maximum median radical concentrations were 4.5×106 cm−3 for OH at noon and 3×108 and 2.0×108 cm−3 for HO2 and RO2, respectively, in the early afternoon. The completeness of the daytime radical measurements made it possible to carry out experimental budget analyses for all radicals (OH, HO2, and RO2) and their sum (ROx). The maximum loss rates for OH, HO2, and RO2 reached values between 10 and 15 ppbv h−1 during the daytime. The largest fraction of this can be attributed to radical interconversion reactions while the real loss rate of ROx remained below 3 ppbv h−1. Within experimental uncertainties, the destruction rates of HO2 and the sum of OH, HO2, and RO2 are balanced by their respective production rates. In case of RO2, the budget could be closed by attributing the missing OH reactivity to unmeasured VOCs. Thus, the presumption of the existence of unmeasured VOCs is supported by RO2 measurements. Although the closure of the RO2 budget is greatly improved by the additional unmeasured VOCs, a significant imbalance in the afternoon remains, indicating a missing RO2 sink. In case of OH, the destruction in the morning is compensated by the quantified OH sources from photolysis (HONO and O3), ozonolysis of alkenes, and OH recycling (HO2+NO). In the afternoon, however, the OH budget indicates a missing OH source of 4 to 6 ppbv h−1. The diurnal variation of the missing OH source shows a similar pattern to that of the missing RO2 sink so that both largely compensate each other in the ROx budget. These observations suggest the existence of a chemical mechanism that converts RO2 to OH without the involvement of NO, increasing the RO2 loss rate during the daytime from 5.3 to 7.4 ppbv h−1 on average. The photochemical net ozone production rate calculated from the reaction of HO2 and RO2 with NO yields a daily integrated amount of 102 ppbv ozone, with daily integrated ROx primary sources being 22 ppbv in this campaign. The produced ozone can be attributed to the oxidation of measured (18 %) and unmeasured (60 %) hydrocarbons, formaldehyde (14 %), and CO (8 %). An even larger integrated net ozone production of 140 ppbv would be calculated from the oxidation rate of VOCs with OH if HO2 and all RO2 radicals react with NO. However, the unknown RO2 loss (evident in the RO2 budget) causes 30 ppbv less ozone production than would be expected from the VOC oxidation rate.