Invest in Best ETF in 2021— “How To” guide the Do-it-yourself way

How to find top ETF to invest

Is there any vehicle other than direct stock investment which is safer and helps you be well-diversified? How to get a list of all possible permutations and combinations of all types of investments and how to rank them in order of your preference: 1 or 2 or 3 year return; volatility (market swings); risk of loss; industry or sector or geographical diversification; and last but not the least, active management by an astute fund manager or just blindly following a stock market index?

This is the part II post in the series of posts on how I came up with a list of best US listed exchange traded funds (ETF). The first post describes the introduction to the process I followed to identify the best available ETFs tradable in US market.

If you have landed on this page accidentally (or intentionally) and would like to see the first part of this Step-by-step Do It Yourself guide, please click here or the article image below.

Briefly reiterating below what I wrote in the earlier post.

I use the excel file I created by downloading ETF data from Fundsupermart.com and after putting it through an extensive filtration process described clearly in the first post, I will now illustrate the long and hard multistage interwoven process taking roughly 6 months to learn and in turn helping you turn into a better investor … phew!! …

Just kidding !!

The process, as I explain, will be far easier. Let me hand-hold you with this Do-It-yourself DIY guide which you can comprehensively read and completely understand in less than 6 hours (not 6 months).

I would still suggest you to see the first part of this Step-by-step Do It Yourself guide by clicking here.

▄▀ Ⓖⓔⓣ ▄▀▄ Ⓢⓔⓣ ▄▀▄ Ⓖⓞ ▀▄

We start with picking up a category+sector+geography combination e.g. Equity-Consumer-US.

  1. Out of the ETFs selected in previous sections (ETF 1.0) for above combination, I carefully evaluate each ETF by:
  • Examining parameters like historical volatility, Lipper & Morningstar ratings, MSCI ESG Analytics ratings, expense ratio, beta, standard deviation, apha etc. from morningstar.com, FT.com or my demo interactive brokers account.
  • Comparing 3 or 5 year performance of automated periodic investment (SIP)
  • Visually observing their 1y — 5y chart on FSM chart center website

2. This way I get a curated list of all best performing ETFs out of the total universe tradeable in Singapore. Next, I compare selected ETFs based on relative strength and correlation to choose the better ETFs out of a few within same category+sector+geography combination.

3. Since my plan is to invest (lumpsum and periodic investment), no short term trade, I further analyse selected ETFs on at least these 2 points:

  • ETFs mirroring my preferred ones while having lower NAV for systematic periodic investment
  • ETFs having lower lot size / minimum units to trade

4. Lastly, I form a portfolio and backtest it before investing through the following methods:

  • download preferred ETFs’ last 5 years (or since inception if less than 5 years old) daily price data from Yahoo finance and run excel simulator to generate the best backtested portfolio.
  • for a visual check, I plot my portfolio on www.tradingview.com e.g. plot a portfolio of a combination of ETFs e.g. (ARKW*2 + VIXM) to see performance over several years.

The websites I used for further analysis are www.etfscreen.com, www.etfdb.com, www.etf.com, www.dqydj.com, markets.ft.com etc. I also used my demo accounts with Interactive Brokers & Saxo markets to check ETF parameters, live spread, tradability, regulatory restriction, limit, margin and so on.

【🅴🆃🅵 🅿🅰🆁🅰🅼🅴🆃🅴🆁🆂 】

For every category+sector+geography combination e.g. Equity-Consumer-US. I inspect all ETFs from the ETF 1.0 list (described in my previous post) on following parameters:

a) Historical volatility vs. 1 year return

Select ETFs with low historical volatility and high 1 year return e.g. if ETF-x has historical volatility =23.2 and 1 year return = 34.8%, and ETF-y has historical volatility =23.8 and 1 year return = 28.2%, I would pick ETF-x over ETF-y.

However, if ETF-y has historical volatility =20.2 and 1 year return = 28.2%, then I would be sceptical choosing ETF-x and would rather put both on radar while continuing to the next steps in examination.

b) Lipper, Morningstar and MSCI ESG Analytics ratings

I prefer ETF with better rating. Within a category+sector+geography combination, I would generally choose the best rated ETFs to come up with max 3 ETFs.

I will pick FDIS, VCR, RTH based on all 3 ratings.

c) Expense ratio

It is the expense incurred in ETF administrative and other operations. More about it is given in the following investopedia article.

Given all parameters are very close to each other, one should prefer a fund / ETF with lower expense ratio. “Lower” here doesn’t necessarily mean that an easy choice can be made based on a difference of 0.1% e.g. xETF and yETF with expense ratio as 0.85% and 0.95%. I mean, there is no set value / rule using which I can select or reject an ETF out of the Equity-Consumer-US pool in ETF 1.0. As per my discretion in this example, I will put both on radar and continue looking at other parameters.

d) Fund ratios

Beta, standard deviation, Sharpe ratio etc.: I feel Beta plays a crucial part in fund selection because it exhibits how a fund moves in tandem with the market. Since I am not a conservative investor, I do not entirely focus on funds with low betas. However, since I am willing to take limited risk for slightly higher returns, I look for beta around 0.8–1.1, which is the case with most investors. I also feel that for periodic investments like SIP (systematic investment plan) for dollar cost averaging, a high beta fund with high returns (sharpe ratio & alpha) is preferable. I will show this in detail in later paragraphs

e) Fund performance

Now how do we determine fund returns? In two ways — annualized returns against benchmark over 1, 3, 5 years, or cumulative return over a set period.

  • Annualized returns against benchmark give you a consistent performer kind of feeling e.g. 1y return against benchmark (e.g. S&P 500) = +15.0%, 2years return against benchmark = +13.5%, 3 years return against benchmark = +14.5% and so on.
  • You can also measure yearly returns e.g. returns in 2018, 2019, 2020. But don’t forget to measure this against a benchmark. Generally if a fund is giving consistent returns over its benchmark, then it is considered a really good find.
  • Cumulative returns over a set period are simply the total return beween start and end date.

Apart from Yahoo Finance and Morningstar, one may see Cumulative returns for any ETF over last 1 week, 1 month… to 5 years on the following website.

Upon clicking the above link, an ETF.com page will load which contains market data, reference data, statistics, comparisons, charts and so on for QQQ.

One may also use www.dqydj.com to see annualized returns, both as a lump sum starting amount and as a Systematic Periodic Investment.

Simply enter the ETF ticker, Investment amount, Start and End dates to calculate lump sum return. For systematic periodic return, the Toggle Advanced button opens up few more options: tick the Periodic Investments checkbox, select the Period and Amount.

Clicking the Calculate Return button will show a chart of total invested value over the selected period. The values in Final Value($) and Annual return(%) are the key ones I note down. I use this Annual return(%) as one more parameter to compare between similar ETFs.

For every category+sector+geography combination, I finally select a few ETFs based on the aforementioned parameters to build an updated excel sheet. I then move on to using other analytical & visual tools to finalize my selection.

ETF parameters I explained above can also be seen at markets.ft.com. However, I simply wanted to share these brilliant websites for research and analysis (as explained in several points above). Had I shared the FT website details in the first line itself, most readers would have skipped the excellent features found on other websites.

This following screenshot shows how the file looks now. I call it “ETF 1.1”.

ETF 1.1 excel sheet with ETFs further filtered and highlighted

Column K & L contain Morningstar and Lipper ratings respectively. Blue highlighted columns M & N contain the 1 year Standard deviation for fund and category respectively while columns O & P contain the 1 year Beta for fund and category. All this information is taken from markets.ft.com.

The last column, named 3yrSIP, is the return as per Systematic Periodic Investment as calculated from www.dqydj.com.

So I selected three ETFs from Equity+Consumer+US combination. However since I have to diversify across categories, sectors, and geographies, I should invest in only one, or max two ETFs. Lets see further how I use Relative Strength and charts to narrow down my selection.

【 🆁🅴🅻🅰🆃🅸🆅🅴 🆂🆃🆁🅴🅽🅶🆃🅷 】

Relative strength (RS) is a measure of the price trend of an instrument compared to another one. It is calculated by taking the price (or returns) of one asset and dividing it by another. E.g. RS of ARKW vs. QQQ is calculated as RS = ARKW/QQQ.

I compare RS using 3 methods:

  1. Calculated using ETFScreen.com
  2. Plotting ETF charts in FSM charts and other websites.
  3. Plotting chart of “ARKW/QQQ” in TradingView.com

This comparison gives me a good opportunity to identify the best available funds out of the lot.

You can view ETF charts on many websites including the most popular Yahoo finance, Morningstar, and the most pervasive Tradingview.com.

I will rather simply use FSM charts for an initial analysis. E.g. for MGK:

FMSOne.com Charts

In this FSM charts screen, you can choose to display chart of a single ETF or multiple ETFs for comparison e.g. I selected MGK and the index S&P 500 here for comparison. By default, all selected ETF will be plotted in the chart section. One can however click on the ETF name at bottom of chart section to hide its display.

This helps to visually inspect the movement of one ETF relative to another to judge the following for a set of ETFs: annual or monthly range, volatility (beta), seasonal price movement, correlation with other ETFs , intensity of up- or down-move with respect to index (alpha) and many such parameters.

ETF.com chart — https://www.etf.com/MGK#overview

I also used other charting websites like charts on ETF.com, TradingView and Yahoo Finance. In ETF.com charts, one can plot Price charts and Total return charts for specific periods while comparing return values with pre-defined indices. A plot of multiple ETFs on a single chart, similar to FSM charts, is also possible. But the agility, smoothness and easy-to-click & load nature of FSM is what appeals to me most.

Plotting a basic ETF strategy or portfolio is really easy in TradingView. I generally plot a ratio of 2 ETFs to get gauge Relative Strength of one vs. the other. Sample screenshots of RS between FDIS vs. RTH and ARKW vs. QQQ are shown below.

ARKW vs QQQ and FDIS vs RTH

It is quite easily visible from the screenshot that ARKW is a better performer, especially lately, while the FDIS generally ranges between 0.45 and 0.39 and has recently had a sudden fall and sharp recovery to move back into the range. This way, one can figure out which is a better ETF in the long run.

One important note: FDIS is an ETF in consumer discretionary segment whereas RTH is an ETF in the consumer staples segment. These two are not comparable. However, FDIS and VCR fall in the consumer discretionary segment. Below you can see another screenshot of FDIS vs VCR on the left and a screenshot of CHIQ (China consumer discretionary) vs CHIS (China consumer staples) on the right.

You can clearly see that FDIS and VCR go hand-in-hand. The recent shoot up of FDIS vs. VCR might suggest that FDIS is a better performing ETF. However, if you notice clearly, the sharp move is just 0.01 points i.e. from 0.25 to 0.26 which is actually insignificant. But why I preferred FDIS over VCR is it’s lower NAV for systematic periodic investment.

Lastly, to reaffirm my selection, I use another brilliant Risk-Return chart from www.ETFScreen.com to boost my analysis further, visually. Its a very simple X-Y chart with the yellow bubble representing the ETF I am researching (MGK in this example) while other blue dots plotting Return vs. volatility for list of ETFs mentioned in the text box. You may change the ETF list in the texbox to research on a list of your preferred ETFs vs. some preferred one.

ETFscreen chart — https://www.etfscreen.com/risk-return-chart.php?s=MGK

This graph clearly shows whether my selected ETF is better or worse than other ETFs I wish to compare. If I can find a better (higher return, lower volatility) ETF as a blue dot, I will rather research a bit more on that while moving my previously selected ETF out of the final list.

After all this analysis, I choose the most promising ones, highlight them in green text in green cell color and now call my list “ETF 1.2”.

So, have I finalized my ETF list yet or is there something else? Almost but not final yet. The last step is to consider my investment appetite, horizon and regularity. My prime reason for creating this list was to keep investing regularly and hold for the next 5 years unscathed by the market moves in near or far term.

My investment appetite (portfolio risk and amount) is based on the following asset allocation:

Asset allocation for ETFs:
10% in Gold
75% in Equity
15% in smart Fixed Income

Note that this is not my asset allocation for my entire wealth but only for this goal of 5 year periodic investment in ETF. Asset allocation for your entire wealth is beyond the scope of this blog post.

Now you may ask the question how to decide between all Gold tracking ETFs: IAU, AAAU, GLD and both the US consumer staple ETF: FDIS & VCR. They are a mirror of each other since they track the same index. My investment allocation and appetite permits me to invest only 10% ($100 out of $1000 monthly) in gold and instead of investing all $100 in just 1 day in a month, I would to split it over several days to capture market movements (automated periodic investment; not trying to time the market) and possibly a better price. So instead of investing in GLD (NAV of US$ 175) I will instead invest in IAU or AAAU (NAV of approximately US$ 18) 4–6 times a month.

Another reason to choose mirror ETFs is the minimum lot size required. A simple example of an illusion in this approach is to choose between the ubiquitous QQQ and a really low priced NASDAQ tracking low priced HKD denominated ETF: 3086 HK. It is priced at 22.38 HK$, roughly equivalent to US$ 3.9. Now if I invest through this ETF, I get a chance to buy 1 quantity on many days in a month or buy say 5–6 quantity every week to fulfill my desire of investing $80–100 monthly in NASDAQ. Its seems brilliant but only in theory since the minimum quantity to buy is not 1 but 200. That makes it an investment of roughly US$ 770. Following is an example of a screenshot of an order from my demo account in interactive brokers.

Final Portfolio based on asset allocation

In the next post I will show you the final ETFs worthy of investment based on all criteria, parameters, evaluation, research and analysis done in this post and the first post.

However before ending this post, I will briefly show you in the few paragraphs below how you can visualize a very basic portfolio or strategy in TradingView. Let’s assume I selected ARKW or XITK (US tech Equity), FPXI (Global Large cap), CHIS (China consumer Equity), VIXM (US volatility), SWAN (Global hedged Fixed income), and GLTR (precious metals).

  1. First a plot of just ARKW:
ETF Portfolio using single ETF

2. Now a kind of a hedged ARKW i.e. a mix of ARKW & VIXM

ETF Portfolio using ARKW and VIXM

3. Now a relative strength plot of a hedged XITK vs. ARKW

Relative strength of an ETF Portfolio using XITK + VIXM vs. just ARKW

4. Finally, a plot of portfolio using ARKW, FPXI, IAU and hedged using option volatility tracking ETF i.e. VIXM.

ETF Portfolio using ARKW, FPXI, IAU and VIXM

Please note: None is these combinations of ETF is any form of recommendation. Please use your own judgement, analysis, research and discretion in creating a portfolio. After all, its your own hard-earned money.

Mind your (co)relationship

One last point — you should be careful on the correlation aspect in your portfolio. A portfolio based on asset allocation should be well diversified. To create such a portfolio, we need to select ETFs based on least correlation between each other. After all there’s no point selecting 5 ETFs highly correlated with each other. It’s like putting all your eggs in one basket. Following screenshot shows an example of a few ETFs and their correlation with each other. These ETFs are from different sectors and geographies. Keeping this in mind ensures I spread my portfolio across sectors, categories and geographies. Otherwise, as an example, if I make a portfolio comprising of QQQ & ARKW in equal ratio forming 40% of my entire portfolio amount (assuming 5 ETFs with equal $$ in each), a minor shock to Tech stocks will give a big jolt to my portfolio.

I created this as a part of a series of investment knowledge posts. All with the sole intention of sharing my knowledge and experience in

  • searching for good quality investment vehicles
  • analyzing and researching on choosing between various options from those
  • finding an easy, scalable and manageable way to invest in those
  • creating and managing my own portfolio while trying to find a balance between investment expectations, regulatory hurdles and reality
  • ensuring the portfolio is in accordance with general asset allocation principles as per my age, investment amount and risk appetite.

Disclaimer: I am not promoting FSM or Interactive Brokers or Saxo markets. At the time of writing this post, I have only 1 Regular Savings Plan with FSM, 1 live Interactive Brokers and 1 demo Saxo markets account. I use my demo accounts for research.

This post is originally published at FiveStepGuide. There, I have written much more content with more examples.

Data Engineer. Investor, blogger (https://fivestepguide.com) quadragenarian, father of 4 year old. I like to share knowledge & experience gained over 20+ years