Poly(thienylene vinylene) (PTV) is a low-bandgap polymer but shows poor performance in both light-emitting and photovoltaic applications. Recently a derivative of PTV with carboxylate substitution, poly(3-carboxylated thienylenevinylene) (P3CTV), has been synthesized and was shown to be fluorescent. The photovoltaic power conversion efficiency based on P3CTV has been found to be much larger than PTV, indicating an intrinsically concomitant relationship between light-emitting and photovoltaic properties. Employing quantum chemistry calculations coupled with our correlation function formalism for excited state decay and optical spectra, we have investigated a series of side-chain substituted PTVs targeting optimal optoelectronic performance. We predict that the carbonyl substituted PTV is a strongly fluorescent polymer with low bandgap, long exciton lifetime, and large spectral overlap between emission and absorption. It is expected that carbonyl PTV is a promising light-emitting and photovoltaic polymer. Methodology wise, we find that (i) our correlation function approach to calculate the optical spectrum has much lower computational scaling with respect to the system size than the conventional method; (ii) the harmonic oscillator approximation for the nonadiabatic decay works better for a large system than for a small system.