Inside the Philippine Stock Exchange — A Data Look at PSEi

Every business news segment in the Philippines mentions the PSEi. "The index closed up 0.8%." "Foreign selling continued for the fifth straight session." These updates are everywhere, but they're always just today's number compared to yesterday's. I wanted to zoom out.

What does a decade of PSEi data actually look like? Which sectors drive the index? Who's buying and who's selling? And how did the market really behave during the COVID crash? I pulled ten years of trading data to find out.

What This Project Covers

The analysis spans 2014 to 2024, using daily price data from Yahoo Finance and PSE reports. It covers the PSEi composite index, the six sector sub-indices (financials, industrial, holding firms, property, services, and mining/oil), and net foreign buying/selling data.

I built rolling return charts, a sector correlation matrix, drawdown analysis, and foreign flow visualizations that show the bigger picture behind the daily noise.

The Motivation

I'm interested in financial data, and the PSE is an underexplored market from a data perspective. Most freely available market analysis focuses on US stocks. Philippine market data is out there, but nobody's really packaged it into an accessible, visual format for people who want to understand the PSE beyond the daily ticker.

I also wanted to test some quantitative finance techniques — rolling returns, drawdown measurement, correlation analysis — on a market that behaves differently from the S&P 500. Emerging markets have their own patterns, and the PSE is a good case study.

Rolling Returns and Correlation Matrices

The main analytical tools here were rolling window calculations. I computed 30-day, 90-day, and 252-day (one trading year) rolling returns for each sector index. This shows not just whether a sector went up or down, but the persistence and volatility of its returns over time.

For the sector correlation matrix, I calculated pairwise Pearson correlations on daily returns across all six sectors. A high correlation means sectors move together; low correlation means they're somewhat independent. I computed this for the full period and also for rolling 90-day windows to see how correlations shift during stress periods (spoiler: everything becomes correlated during a crash).

Foreign flow analysis was simpler technically but required cleaning. PSE reports net foreign buying/selling daily, but the data has gaps on holidays and occasional reporting inconsistencies. I aggregated to weekly and monthly levels to smooth out the noise and computed cumulative foreign flows to show the long-term direction.

What the Market Data Shows

Property and financials dominate the PSEi by market capitalization. These two sectors account for a disproportionate share of the index's movement. When Ayala Land or SM move, the entire index moves. That concentration is both a feature and a risk.

The foreign selling trend is one of the most striking patterns. Foreign investors have been consistent net sellers for years. The money has been flowing out, not in. This matches a broader trend of foreign capital leaving emerging markets for higher-yielding US assets, but the persistence in the Philippine market is notable.

• The PSEi peaked near 9,000 in early 2018 and hasn't reclaimed that level
• The COVID crash in March 2020 saw the index drop roughly 35% in weeks
• Recovery was uneven — financials and property bounced faster than mining and industrial
• Sector correlations spike to near 1.0 during market panics, reducing diversification benefits
• The holding firms sector (SM, Ayala, JG Summit) effectively mirrors the composite index
• Daily trading volume has declined over the decade, suggesting lower retail participation

The COVID crash and recovery is the most dramatic segment in the data. The speed of the selloff was extraordinary — decades of market history suggest crashes that steep should take months, not weeks. The recovery was also faster than historical patterns would predict, though the PSEi never fully returned to its pre-pandemic highs during the analysis period.

What I Took Away

Working with Philippine market data reinforced something I already suspected: the PSE is a concentrated market dominated by a handful of conglomerates. The "index" is really a story about SM, Ayala, and a few others. That's not necessarily bad — it's just something investors should be aware of.

If I extend this project, I'd add individual stock-level analysis and look at how corporate earnings announcements affect price movements. I'd also like to incorporate BSP (central bank) rate decisions to see how monetary policy transmits into stock prices. But even at the sector level, the patterns are clear and, I think, useful for anyone trying to understand how the Philippine market actually works.