Using ChatGPT To Review Our Bedokian Portfolio

Out of curiosity, we asked ChatGPT (free version) to review our Bedokian Portfolio, after reading up online on what other investors had done. With the exception of removing our holdings list, the answers were “as is” from ChatGPT (though from the replies you could catch a glimpse of what we are holding). 

Picture generated by ChatGPT

The prompted question was: Please analyse the following portfolio. Numbers denoting quantity of securities of the ticker. Only the counter names and number of securities were provided to ChatGPT.

I will give my take at the end of the post. Meanwhile, enjoy the ChatGPT review!

---

Below is a structured, portfolio-level analysis based on your holdings and quantities. I’ll focus on asset allocation, concentration, risk profile, income vs growth balance, and key strengths/risks, then end with practical suggestions(not advice).

---

 

1. High-Level Asset Allocation (by asset type, not market value)

Because prices aren’t provided, this is qualitative weighting based on number of holdings and intent.

🏢 REITs & Property (Very Heavy)

~30–40%+ of portfolio (likely more by value)

Observations

• Strong Singapore REIT bias
• Diversified across industrial, retail, office, and data centres
• Heavy reliance on interest-rate direction and refinancing risk

Pros

• Stable income
• Inflation-linked rents (partial hedge)
• Familiar regulatory environment (SG)

Risks

• Rate-sensitive
• Geographic concentration (Singapore-centric)
• Structural office risk (Suntec, Lendlease)

---

📈 Equities – Growth & Quality

~25–35%

Observations

• Strong tilt to US mega-cap tech
• Nvidia + ASML = high semiconductor cycle exposure
• Berkshire adds stability and capital discipline

Pros

• Long-term growth engine
• Currency diversification (USD exposure)
• High-quality franchises

Risks

• Tech valuation cyclicality
• Concentration risk (AI/semiconductor theme)

---

🇸🇬 Singapore Equities (Non-REIT)

~10–15%

Pros

• Defensive, cash-generative businesses
• Dividends and balance-sheet strength

Risks

• Slower growth
• Small-cap liquidity (YHI)

---

💵 Fixed Income / Bonds (Very Strong)

~25–30%

Observations

• Excellent capital preservation sleeve
• Laddered duration via SSBs
• High credit quality overall

Pros

• Stability
• Income buffer during equity drawdowns
• Strong SGD base

Risks

• Reinvestment risk if rates fall
• Lower real returns long term

---

🌱 Thematic / Satellite ETFs

~5–8%

Pros

• Optionality and upside themes
• China exposure via A-shares (policy levered)

Risks

• Volatility
• Policy and regulatory risk (China, clean energy)

---

🪙 Commodities / Hard Assets

~3–5%

Pros

• Inflation hedge
• Crisis insurance

Risks

• No yield
• Silver volatility

---

2. Portfolio Characteristics Summary

Dimension	Assessment
Income Generation	⭐⭐⭐⭐⭐ Very strong
Capital Preservation	⭐⭐⭐⭐½
Growth Potential	⭐⭐⭐⭐
Geographic Diversification	⭐⭐⭐½
Interest Rate Sensitivity	⚠️ High
Currency Diversification	Moderate (SGD + USD heavy)

---

3. Key Concentration Risks to Watch

🔴 1. Singapore REIT + SGD Risk

• REITs + SGD bonds + SG equities = macro concentration
• If SG property or rates move adversely, multiple assets correlate

🔴 2. Interest Rate Sensitivity

• REITs + bonds move inversely to yields
• A sharp rate spike would pressure both income sleeves

🔴 3. AI / Semiconductor Cycle

• Nvidia + ASML + S&P 500 exposure
• Fantastic upside, but cyclical drawdowns can be severe

---

4. Strengths of This Portfolio

✔ Well thought-out income + growth barbell
✔ Strong use of ETFs for efficiency
✔ Conservative bond allocation cushions volatility
✔ Real assets + equities + bonds = true multi-asset structure
✔ Suitable for long-term wealth preservation with income

---

5. Optional Refinement Ideas (Not Advice)

You may consider:

1. Reducing REIT concentration over time and reallocating to:
• Global infrastructure
• Global dividend ETFs
2. Adding non-US developed equities beyond EAFE (e.g. factor or dividend tilt)
3. Monitoring duration risk if rates fall sharply (bond reinvestment strategy)
4. Rebalancing AI exposure periodically due to valuation swings

---

Bottom Line

This is a mature, income-anchored, institution-style portfolio with:

• Strong downside protection
• Sensible growth engines
• Some macro concentration risk that is manageable with rebalancing

---

The Bedokian’s Take

We were somewhat surprised by ChatGPT’s analysis, yet we also found the results fairly consistent with our portfolio construction. Reviewing point 2, Portfolio Characteristics Summary, ChatGPT awarded a 5-star rating for income generation, a 4½-star rating for capital preservation, and a 4-star rating for growth potential. When considering the portfolio strengths mentioned in point 4, it became clear that our current Bedokian Portfolio aligns well with our expectations.

Looking at the drawbacks, the main issue was the portfolio's high sensitivity to interest rates. This had been anticipated since the Bedokian Portfolio allocates a significant portion (20% to 40%) to REITs. Although REITs serve as valuable sources of income, they are notably sensitive to interest rate changes—a trend we have observed from early 2022 to recent months.

In addition to the Singapore Dollar bias that ChatGPT identified (without prior information about its focus on Singapore-based investors), another risk highlighted was the considerable exposure to artificial intelligence and semiconductor sectors. This aspect of our Bedokian Portfolio includes stocks like ASML, Nvidia, Apple, Alphabet, and is further influenced by the technology-heavy S&P500 index. Such a composition was purposefully chosen for growth, as explained in this post, and to capture extra alpha as detailed in my previous blog entry.

Overall, this exercise was insightful and confirmed several objectives we set for the Bedokian Portfolio. We may consider providing more data to ChatGPT and other AI tools to gain further views on our investment approach.

Generation (Of) Alpha

Though the title may sound similar, this is not about a post on people who were born in between the entirety of 2010s and early 2020s. Rather, we are exploring the financial parlance for the word alpha, which is the excess above market returns.

Picture generated by ChatGPT

Using one of the U.S. markets’ indices as an example, the S&P 500, it returned around 17% for the whole of 2025. Meanwhile, Nvidia, a constituent of the S&P 500, returned close to 36% for the same period. Thus, the alpha generated by Nvidia was 36 – 17 = 19%.

Alpha is relative to a selected benchmark index, so choosing the right benchmark is important for accurate interpretation. In the above example, though the S&P 500 was used, other major indices like the Dow Jones 30 and NASDAQ could also be utilised as they are the main representations of the U.S. market in general (which Nvidia is part of as well). Conversely, although the Russell 2000 is a major index, it is not appropriate to compare it with Nvidia, a large-cap counter, for alpha as the former is the primary benchmark for small-cap equities.

On top of individual companies, alpha can be calculated on whole investment portfolios, too, in gauging performance. For instance, a portfolio made up entirely of U.S. equities could be compared with the S&P 500 or Dow Jones, and a portfolio of Singapore counters could be used with the Straits Times Index. Sometimes a mixed local-foreign portfolio is weighed up against relevant indices. Ultimately, it is up to the individual investor on which to apply on, provided it is relevant and reasonable.

Generating alpha on a portfolio level requires the employment of active investing, be it via individual counter selection and/or going into active funds. On a multi-portfolio level (e.g. the Portfolio Multiverse concept), on top of what was suggested in the previous sentence, the Trading Portfolio is a way to do so, in which it may be a possible option for passive investors with index funds in the main portfolio. Another way for passive investors is either to adopt a core-satellite structure (core of passive index funds with satellite of active funds/individual securities) or to create another portfolio for active investing. The combinations are varied and flexible.

For our Bedokian Portfolio, we adopted the core-satellite methodology, and some of our satellite counters are part of index exchange traded funds (ETFs) in our core; while this may be viewed as an overlap (e.g. “why buy Nvidia when you already have the S&P 500 ETF?”), the securities were selected based on different merits. For this case, we have the S&P 500 for U.S. market exposure, but we are vested in Nvidia for the artificial intelligence growth potential with its chips, subject to our own weightage limit for individual company counter (i.e., the 12% limit). This is one of our ways in the generation of portfolio alpha.

Disclosure

The Bedokian is vested in the S&P 500 via the SPY ETF, and in Nvidia.

Disclaimer

Bob’s Portfolio Update

Picture generated by Meta AI

In my post just before the year ended, I had mentioned that Bob, our hypothetical Bedokian Portfolio investor, would rebalance on 2 Jan 2026 using the SGD 5,000 cash injection.

After the rebalance, Bob’s Bedokian Portfolio status is shown in the link here.

Going forward, I would provide the occasional snapshot of Bob’s portfolio in this page.

Some of you might have recalled that I would be changing Bob’s REIT ETF this round as per what I had stated in my last post. However, the current REIT ETF is in its cum-dividend period, while the targeted REIT ETF to change over to had just gone into ex-dividend. Hence, in not losing out on the distribution amount, it would be prudent to wait it out and convert over later.

2025 Review, 2026 Preview And Bob

It is that time of the year again where we will go through what had happened the past year, what may happen the next year, and some updates on our hypothetical investor Bob.

Picture generated by ChatGPT

2025 Review

Going through the usual markets which I cover annually, the S&P500 had returned +17.86% year-to-date (YTD) while, surprisingly, the Straits Times Index (STI) returned higher at +22.41%, and reaching an all-time high of above 4,600 points as at the close of Christmas Eve.

The STI, often seen as a laggard and bypassed by some local investors, saw a period of happy days over the past year, to which in my view was partially attributed to capital flowing from other parts of the world to our deemed safe haven markets.

The best story for this year comes not from equities, properties or bonds, but from gold and silver, which both have a place in The Bedokian Portfolio. The news of debasement of the United States dollar coupled with the increased use of silver in industrial (and political) use resulted in a spike of the two precious metals’ prices mainly in the second half of the year, with gold and silver having +70.64%¹ and +143.01%¹ YTD respectively, beating the aforementioned equities indices.

HACK, IPAY & ICLN

Since last year, I have included a section in my year-end posts highlighting the three sectors/industries: cybersecurity, payment solutions, and clean energy, represented by the exchange traded funds (ETFs) HACK, IPAY, and ICLN, respectively, in which we are vested in. The table below (Fig. 1) shows the ETFs’ YTD performance for the whole of 2025 (up till 24 Dec 2025) and the compounded annual growth rate (CAGR) from 1 Jan 2018 to 30 Nov 2025 (nominal and inflation adjusted) since I had espoused them in 2017.

 

ETF	2025 YTD (till 24 Dec 2025)	Nominal CAGR (1 Jan 2018 to 30 Nov 2025)2	Inflation Adjusted CAGR (1 Jan 2018 to 30 Nov 2025)2
HACK	+9.99%	+13.51%	+9.74%
IPAY	-8.17%	+5.43%	+1.93%
ICLN	+45.06%	+9.37%	+5.73%

 

Fig.1: YTD and CAGR nominal and inflation adjusted returns of HACK, IPAY and ICLN

2026 Preview

Amongst the boomtown happenings, we also heard about economists and analysts projecting a market downturn next year, amplified by news of layoffs (happened and projected) and increasing unemployment numbers. The constant push and pull of conciliatory and retaliatory trade measures between the two largest economies, the United States and China, brought about more uncertainty that ironically in my opinion, the resulting rise of our local stock market.

Whether good times or bad are forecasted, it is important to note that we cannot really tell what lies ahead, but it is equally important to stay invested and diversified; diversification after all is the first line of defence against market uncertainty.

Bob

I had just realised Bob’s portfolio did not show up as it should be on the blog page, hence going forward I will transfer the portfolio and post it as a static table form after rebalancing on 2 Jan 2026 with an SGD 5,000 cash injection.

Anyway, Bob’s portfolio value is now close to SGD 160,000 and had collected almost SGD 4,600 in dividends. Also, I would be changing Bob’s REIT ETF to another after considering the other ETF’s fund size and liquidity, so stay tuned after 2 Jan 2026 for the reveal of Bob’s portfolio.

Onward to 2026!

Disclosure

The Bedokian is vested in HACK, IPAY and ICLN.

Disclaimer

All figures are from Yahoo Finance as of 25 Dec 2025 unless otherwise stated.

1 – TradingView (accessed 25 Dec 2025)

2 – Portfolio Visualizer, HACK, IPAY and ICLN between 1 Jan 2018 to 30 Nov 2025 (accessed 25 Dec 2025) 

Is Such Knowledge Really Necessary?

 A few days ago, someone had posed me this question:

“Is it necessary to know all these before starting to invest?”

Picture generated by Meta AI

By “all these”, the person was referring to a blogpost I made back in 2021 (link here), where I had introduced a structured and holistic guideline to learn investing. An acquaintance of mine, after reading the post, then asked the above question. The viewpoint provided was that such knowledge was deemed too overwhelming and not suitable for people who may not have the aptitude to grasp the concepts.

Honestly, this was not the first time I was queried on it, so I am writing this post providing my viewpoints and use the link as my answer in case anyone asking me again in future.

No doubt it is a daunting task to go through the structured learning guideline on how to go about investing, in my opinion it is important to know what one is getting into. Whenever I embark on a learning journey on a topic, my preference is to know the underlying concepts, or at least the basic understanding of the whole works, and with it, further related information would be easier to catch on to and eventually the learning would become more gradual.

Besides knowing what one is doing on the investment front, the possession of the know-how also serves as a bulwark against fishy investment/trading schemes. Oftentimes there were news reports of people falling for investing frauds which, upon critical thinking and questioning, would have been avoided.

It is not difficult to start learning on investing, for it is akin to learning other new skills and hobbies like cycling and pickleball. The availability of generative artificial intelligence tools like ChatGPT and Gemini makes learning easier nowadays (though some caution and basic fact finding is still needed for this). The choice of learning is there, whether one wants to have the effort to take it up.

Let It Go

Picture generated by Meta AI

We recently sold a dividend stock at a loss. Initially, we bought it in July 2017 after reviewing its strong financials and solid fundamentals (such as price-to-book ratio, gearing, and revenue). When COVID hit in 2020, we purchased more shares as prices dropped, anticipating a rebound once the crisis passed—especially since the company is connected to tourism. The stock did recover somewhat, but after 2022 it began to decline again and has stayed weak since. Additionally, the dividend yield kept decreasing each year, eventually falling near to the 10-year annual inflation average of 1.75%¹.

In total, we incurred a -32.6% loss based on our entry and exit prices, made worse by the stock’s wide spread and low liquidity, which meant selling at the bid price. Even after accounting for the dividends we received, the overall outcome was still negative at -21.3%. The good thing was the company share represented only at 0.4% of our total Bedokian Portfolio value.

Cutting losses can be tough for investors, but it is often necessary to free up capital for better opportunities rather than letting funds stagnate. Potential price rebounds or dividend increases are not guaranteed. Instead of viewing one’s investment as hard-earned money lost, consider it as strategic capital to be redeployed effectively. This is one of the true marks of a rational investor.

Related post

Are You Mentally Prepared For Investing?

1 – MAS Core Inflation, 2015 to 2024. Monetary Authority of Singapore. 

Rebalancing: Adding Bond ETFs

While many investors and traders are focused on artificial intelligence counters, precious metals like gold and silver, or cryptocurrencies set for recovery, we are going contrarian and turning our attention elsewhere: bonds, specifically bond exchange traded funds (ETFs).

Picture generated by Meta AI

Although we remain opportunistic with other asset classes and sectors, taking positions when prices are favourable, the main portion of our upcoming portfolio additions will be allocated to bond ETFs. This may seem surprising, but from a holistic portfolio perspective, this contrarian approach plays an important role in rebalancing, which is essential for diversification.

Currently, bonds make up 12.7% of our portfolio. As previously discussed here, our target allocation for bonds is 15%, so we are aiming to increase our holdings nearer to that percentage figure.

We prefer bond ETFs over individual bonds because they offer easier management, greater diversification thanks to exposure to different issuers, better liquidity since ETFs are easily traded on markets, and lower transaction costs by dealing with one ETF instead of multiple individual bonds.

That said, we also hold some individual bonds such as Singapore Savings Bonds, which we intend to keep until maturity in order to benefit from their higher interest rates in specific tranches.

Disclaimer