Learn from the experts using AI

I occasionally browse Jishi Lu, following some experts whose returns consistently outperform the index, such as San Ceng Ge. However, I can’t really understand the author’s updated rebalancing records, so I try to use AI to help me analyze. The approach is to instruct openclaw to download and package all the author’s posts, then feed them into notebooklm for analysis. I then ask notebooklm questions as if it were the author, which seems to yield some insights.

Q1: Please provide a detailed introduction to your investment strategy, and how beginners can learn and apply it.

“San Ceng Ge”'s investment strategy has been tested over more than ten years of market cycles, forming a complete system, mechanical in operation, suitable for long-term investors with low-frequency trading. Its core aims to address the two major challenges faced by ordinary investors: “timing the market” and “stock selection.” The author modestly states that this method is more suitable for “those with low skill, slow reactions, and not pursuing high returns.”

Below is a detailed introduction to this investment strategy and some learning advice for beginners:

1. Detailed Explanation of the Investment Strategy

1. Asset Allocation and Dynamic Rebalancing (Solving the “Market Timing” Problem)

• Classic 70/30 Stock-Bond Ratio: Maintain a fully invested position at all times, with a baseline allocation of 70% equities (aggressive) and 30% fixed income (conservative).
• 2% Dynamic Adjustment Threshold: When the proportion of a asset class deviates from the baseline by more than 2%, rebalancing is triggered (a “foolproof” operation). For example, if equities rise to 72%, sell 2% of equities to buy fixed income; if it drops to 68%, sell some fixed income to buy equities. This mechanism forces investors to “buy high and sell low,” abandoning market top/bottom predictions and timing.
• Evolving Allocation: As market conditions change, the author further refines the baseline allocations by the end of 2025 to: 35% A-shares funds, 35% US stocks funds (Nasdaq and S&P), 15% bonds funds, and 15% gold ETFs, again with 2% thresholds for dynamic adjustment.

2. Security Selection and Parallel Rebalancing (Solving the “Stock Picking” Problem)

• Focus on Closed-End Funds (CEFs): The author mainly trades CEFs or fixed-open funds in the market, emphasizing their “Net Asset Value (NAV) growth” (offensive leg) and “annualized discount” (defensive shield).
• 4-week/10-week Net Growth Rotation Method: Buy targets based on top rankings of “net NAV growth + annualized discount” over the past four weeks; sell targets based on poor performance in “net NAV growth + discount” over the past ten weeks. The core idea is “discard the weak and keep the strong.” As long as the worst funds are not held long-term, the strategy is successful.
• Spread-out Defensive Approach: Recently, due to significant narrowing of discounts and low trading volume (poor liquidity), friction costs have increased. The author has gradually reduced frequent rotations, shifting toward a “spread-out” (diversified) approach, also introducing US ETF and bank convertible bonds for substitution and observation.

2. How Beginners Can Learn and Use This Strategy

For those interested in learning this approach, the author offers the following core advice:

1. Understand the Principles, Don’t Blindly “Copy”
The author repeatedly emphasizes that his returns are only moderate in the forum. Blindly copying specific holdings is not wise. Copying can be a shortcut, but only if you “understand and agree” with the underlying logic of the strategy. Otherwise, during market fluctuations or losses, your mindset will suffer greatly.

2. Do Your Own Homework and Build Data Tracking Sheets
Beginners are advised to think independently and create data tables in Excel. Simply record the net value and trading prices of market funds every weekend, then calculate “discount rate,” “4-week net growth,” and “10-week net growth,” and sort accordingly. Doing data manually helps clarify the causes and effects, making it easier to improve based on personal circumstances.

3. Stick to the “Idle Money Investment” Bottom Line
Invest only with idle funds that are not needed in the short term. Using idle money combined with the stock-bond dynamic balance strategy ensures that, regardless of market volatility, you won’t panic. During major declines, you can even “play ostrich” and lie flat.

4. Abandon Predictions, Embrace “Foolproof” Operations
Recognize that ordinary people cannot accurately predict stock market rises and falls. Don’t overestimate your ability. Set clear rules (like 70/30 ratio and 2% thresholds), then execute mechanically—no need to watch the market daily; review on weekends. Accept gains and losses calmly, and only aim for the average returns within the rules.

Q2: How are specific targets selected?

Based on “San Ceng Ge”'s actual trading records and sharing, he chooses specific targets (mainly market-in closed-end funds) with a very mechanical and clear set of rules, which have evolved over time according to market conditions. The specific method and logic are as follows:

1. Core Selection Criteria: Net Growth + Discount

The author likens closed-end funds to having “two legs”:

• Net Asset Value (NAV) growth (offensive leg): Represents the fund manager’s stock-picking ability and the fund’s offensive capacity.
• Annualized discount (defensive shield): Represents the safety cushion and defensive strength brought by the closed period.
  The ideal targets are those with continuous NAV growth and either expanding or stable discounts.

2. Specific Buy/Sell Rules: Four-week Entry, Ten-week Exit

To determine when to buy or sell, the author establishes a time-based observation cycle:

• Buy signals (look at four weeks): Focus on the fund’s recent four-week NAV growth, combined with ranking by annualized discount. The top-ranked funds are preferred for purchase.
• Sell signals (look at ten weeks): Focus on the recent ten-week NAV growth. If a fund’s four-week and ten-week NAV growth are both weak, and its discount has shrunk, and it ranks below the median among CEFs, it is sold and replaced.
• Discard the weak, keep the strong: The core idea is “fuzzy operation”—not necessarily selecting the best-performing funds, but replacing those with deteriorating indicators.

3. Analogy for the Selection Logic: “Walking the Dog”

The author uses the “walking the dog” analogy to explain the internal logic of profiting from discounts:
The fund’s NAV is like the “owner,” and the trading price is like the “dog.” The dog may run ahead or lag behind, but ultimately, it will return to the owner (i.e., at maturity or open-end, discounts will zero out).
When the “owner” (NAV) runs fast (NAV growth), but the “dog” (price) lags behind, causing the discount to widen, it indicates that the price will eventually catch up—an excellent buying opportunity.

4. Tools for Selection: Manual Excel Tracking

The author does not rely on sophisticated software but manually inputs NAV and trading prices into Excel every weekend, calculates “4-week net growth,” “10-week net growth,” and “annualized discount,” then sorts and analyzes. He believes that doing it himself ensures understanding of the cause-and-effect.

5. Evolved Strategies: Spread-out and Diversification

As the market develops, the author adjusts his approach:

• Reduce rotation frequency, shift to “spread-out” (diversified) holdings: Due to narrowing discounts and low liquidity, frequent trading incurs high friction costs. He now disperses holdings across many funds to reduce liquidity risk.
• Include conservative assets: Besides early pure bond funds, he introduces bank convertible bonds (valued for low price, safety, and low volatility) and gold ETFs as alternative risk diversifiers.
• Cross-market assets: To diversify further, he allocates to overseas markets, such as Nasdaq ETFs/LOFs (representing global tech stocks) and S&P 500 ETFs. Occasionally, he exploits premium differences between Nasdaq ETFs and LOFs for arbitrage, earning small profits.