SMN Digest October 2021

1. Implementing the Modified Nelson-Siegel-Svennson Model in Constructing the Philippine Yield Curve

Armin Paul D. Allado and Dr. Ruel V. Maningas

The main goal of this paper is to construct the Philippine yieldcurve using the Modified Nelson-Siegel-Svennson (NSS)model, a parametric method commonly used by economists to understand the dynamics of the yield curve. The NSS model incorporates six parameters to determine the location and magnitude of the curvature of the yield curve. Constructing the theoretical yield curve involves determining the optimal parameter values that will minimize the residual between fitted and actual yields. 

The model takes as input the real-time trading data for all bonds transacted on a particular date and finds a smooth curve that best fits the observed yields using the six parameters in the NSS equation. Moreover, the model can identify which bonds are undervalued or overvalued based on the best fit curve. Furthermore, the optimal parameters of the model were analyzed to understand how they affect the best fit estimation.

2. Triple Bottom Line Predictors of SMEs’ Trade Credit

Marjory Uy Legaspi

This research aims to determine the significant effect of triple bottom line predictors to trade credit. This is essential in order to help SMEs identify predictors in granting trade credit to credit-worthy customers. The triple bottom line predictors namely, social, environmental and economic dimensions yielded a significant effect to trade credit using the multiple regression model. This research exuded a holistic approach in evaluating customers because it considers Edward Freeman’s theoretical framework and Elkington’s sustainable TBL model. Furthermore, this research used the Sequential Mixed method. Causal research design was also utilized. The participants of the study were SME entrepreneurs, finance heads, credit department heads and business owners’ representatives. Primary data were initially collected through survey questionnaires to 268 samples and later made interviews to 9 respondents of the 268 sample SME owners or their representative credit department heads who garnered the highest, lowest and average scores per dimension. This research found out that the triple bottom line – social, environmental and economic dimensions significantly affected trade credit using multi-linear regression analysis. In addition, each element per dimension on the average was considered by the SME creditors whenever they grant credit to customers. Finally, it was proven in this research that the triple bottom line dimensions can now be utilized as predictors of trade credit. There is now empirical evidence supporting the TBL model of Elkington and the holistic business perspective of Edward Freeman on Stakeholder’s Theory.

3. Travel Package Recommendation System Using Topic Modeling Approach

Marianne P. Vitug, MABA

The goal of the study is to build a recommender system that will incorporate important factors in designing a travel package – the inherent attributes of tourist destinations, and their distance from other tourist destinations. This project sourced its primary data from reviews from Tripadvisor collected using the package rvest from R. 37 current tourist destinations were included in the data gathering with 50 reviews gathered for each location. 

Distance matrices using Geodesic distance calculation and driving distance via Google API were gathered for this project. These were used as a penalizing factor to produce the hierarchy of the final set of recommendations. The Geodesic distances between the various tourist destinations were gathered using R ggmap and lmap libraries. For the Google driving distance, Google Cloud platform’s Distance Matrix API registration had been necessary to get the key that was used for the R code using gmapdistance library. The Google Distance Matrix API gave travel distance based on the recommended route for a supplied matrix of origins (start) and destinations (end point). 

For the modeling technique, since the data had characteristics that were unsuitable for the algorithms commonly used in recommender systems, topic modeling was used as an alternative method of extracting the intrinsic features of both the tourists and the locations. Latent Dirichlet Allocation (LDA) and Bidirectional Encoder Representations from Transformers (BERT) were used for topic modeling. 

The results were evaluated using a mix of eyeballing on top N words and intrinsic evaluation metrics through topics interpretability. Aside from this, the final list of recommendations was sent to Ark Travel’s President and the head of local tour operations for evaluation and successfully fitted their requirements. The model was able to produce recommendations which are deemed acceptable based on these criteria. The solution made using this recommender system can help not only the main stakeholders— the travelers and the travel agency, but also the business owners on less popular or just-emerging tourist destinations since they can also be recommended as long as they are part of the dataset.

All articles published in SMN Digest are indexed in EBSCO Information Services and may be accessed in full text through Philippine E-Journals.