Measurements of cosmic-ray (CR) positron fractions by PAMELA and other experiments have found an excess above 10 GeV relative to the standard predictions for secondary production in the interstellar medium. Although the excess has been mainly suggested to arise from some primary sources of positrons (such as pulsars and or annihilating dark matter particles), the almost constant flux ratio of
argues for an alternative possibility that the excess positrons and antiprotons up to the highest energies are secondary products generated in hadronic interactions. Recently, Yang & Aharonian revisited this possibility by assuming the presence of an additional population of CR nuclei sources. Here we examine this secondary product scenario using the DRAGON code, where the radiative loss of positrons is taken into account consistently. We confirm that the CR proton spectrum and the antiproton data can be explained by assuming the presence of an additional population of CR sources. However, the corresponding positron spectrum deviates from the measured data significantly above 100 GeV due to strong radiative cooling. This suggests that although hadronic interactions can explain the antiproton data, the corresponding secondary positron flux is still not enough to account for the Alpha Magnetic Spectrometer data. Hence the contribution from some primary positron sources, such as pulsars or dark matter, is nonnegligible.