This paper investigates how individuals' product choices are influenced by the product choices of their connected others and how the influence mechanism may differ for fashion- versus technology-related products. We conduct a novel field experiment to induce and observe choice interdependence in a closed social network. In our experiment, we conceptualize individuals' choices to be driven by multiattribute utilities, and we measure their initial attribute preferences prior to observing their choice interdependence and collecting network information. These design elements help alleviate concerns in identifying social interaction effects from other confounds. Given that we have complete information on choices and their sequence, we use a discrete-time Markov chain model. Nonetheless, we also use a Markov random field (MRF) model as an alternative when the information on choice sequence is missing. We find significant social interaction effects. Our findings show that whereas experts exert asymmetrically greater influence on a technology-related product, popular individuals exert greater influence on a fashion-related product. In addition, we find choices made by early decision makers to be more influential than choices made later for the technology-related product. Finally, using the MRF with snapshot data can also provide good out-of-sample predictions for a technology-related product.