Genomic selection involves using dense markers across the entire genome to estimate the breeding value of selection candidates for quantitative traits. This paper discusses the prediction of breeding values based on a linear combination of markers, where the best estimate of each marker's effect is the conditional expectation of the effect given the data. The accuracy of the estimated breeding value (EBV) can approach 1.0 but requires a large amount of data. An alternative model is explored where only some markers have non-zero effects, following a reflected exponential distribution, which results in higher accuracy. The paper also examines the long-term response to genomic selection, which can decline over generations, and proposes a method to optimize the selection index to maximize this response by varying the weight given to each marker based on its frequency. The introduction highlights the limitations of traditional selection methods and the potential of genomic selection, particularly with the advent of high-throughput genotyping technologies. The paper then delves into the calculation of EBVs from genomic data and the expected long-term response to genomic selection.Genomic selection involves using dense markers across the entire genome to estimate the breeding value of selection candidates for quantitative traits. This paper discusses the prediction of breeding values based on a linear combination of markers, where the best estimate of each marker's effect is the conditional expectation of the effect given the data. The accuracy of the estimated breeding value (EBV) can approach 1.0 but requires a large amount of data. An alternative model is explored where only some markers have non-zero effects, following a reflected exponential distribution, which results in higher accuracy. The paper also examines the long-term response to genomic selection, which can decline over generations, and proposes a method to optimize the selection index to maximize this response by varying the weight given to each marker based on its frequency. The introduction highlights the limitations of traditional selection methods and the potential of genomic selection, particularly with the advent of high-throughput genotyping technologies. The paper then delves into the calculation of EBVs from genomic data and the expected long-term response to genomic selection.