making (a little) money off interest | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Making Real Money in FX 9 replies

Hedging and making Interest 5 replies

Making money in spite of fundemental logic (as we may understand it) 5 replies

Being Right and Making Money Are Not Equivalent, by Dr. Van K Tharp 2 replies

Journal Factory posters - how much money are you making? 17 replies

Attachments: making (a little) money off interest
Exit Attachments

making (a little) money off interest

  • #21
  • Edited Aug 26, 2006 3:27am Aug 25, 2006 9:10pm | Edited Aug 26, 2006 3:27am
Hello to all,

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

I’m trying to find a calculator that will help me hedge correlated currency pairs for profit on interest ie: GBP/USD – USD/CHF and GBP/JPY – CHF/JPY. The calculator would have to be able to balance the unit size of each currency in the hedge. Example right now GBP/USD – USD/CHF is at about three to four ratio P/L is 1 for GBP/USD and 0.8075 for USD/CHF. It would also have to have an input so that the account size is taken into consideration when buying the units size. The calculator would also have to be able to account for currency pair drawdown in the hedge which could possibly cause a margin call. Example if our currency hedge pairs swings 500 pips and the hedging currency only swings 50 pips this could possibly trigger a margin call if the account size is not large enough. Like isufreak24 I’m looking to do all this within one account. Does anybody have any suggestions for sites.
Attached Images
Go in on strategy, come out as planned. Trade like ICE
  • #22
  • Aug 25, 2006 9:42pm Aug 25, 2006 9:42pm
Quoting mrmikal
Witch...we're on the same page. You need to have a high ratio of money to lots in order to withstand a dramatic drawdown.
I looked at this strategy in depth about a month ago or so. One of the things I did was calculated the average weekly move of the GBPJPY. I wanted to know how much money I needed to have in reserve (and therefore, how many lots to trade) so that I could limit transfers to 1-2 times per week. Then I factored into my spreadsheet the cost of doing transfers every few days or so.

So, here are some numbers. Remember, these are just sample numbers to give you an idea. (Also, my calculations are a month old, so the numbers are not perfect.)

Assumptions:
- Use a mini account with at least 200:1 leverage
- Trade GBPJPY with about 10 pip spread.
- One account that pays, the other is interest free

Trade:
- Start with $40,000. Split between the two accounts.
- 5 lots (50 minis) on each side
- Margin: 4661
- Available margin: 15338
- Max negative pips supported with available margin: -351. This should be enough to limit transfers to about once a week, maybe a little more.
- Transfers per month: 5 (assumes you close trade on both sides, transfer funds, then reopen trades.)
- Your net gain if GBPJPY is paying $20 per lot per day is about $2300 per month.
- This would equal a 5.7% MONTHLY gain on your investment of $40,000. It would also be completely risk free, as far as I could tell. So, your looking at about 70% gain.

You can probably trim costs further by finding accounts at 400:1. You could also, instead of keeping money in the accounts, push the limit on margin and keep an extra amount reserved in a bank account. Then, rather than closing the trade, you just transfer from your bank account to the account experiencing the drawdown (the other account would be doing fine.)

I found this a viable, safe method. The only problem is that I, in good conscience, could not ask for an interest free account on religious grounds. It goes against my integrity. Alas.

stockwet
  • #23
  • Aug 25, 2006 9:45pm Aug 25, 2006 9:45pm
Sorry, it actually ends up being a bit more than 70% annually if you reinvest your returns and get the benefits of compounding.
  • #24
  • Aug 25, 2006 9:49pm Aug 25, 2006 9:49pm
I don't think it matters... for instance neuimex offers what it advertises as "islamic swap free" accounts but I think they do that just for advertising... you have to pay a commission to trade swap free accounts with them.

I think it is 5$ per lot, I can't remember.

Quoting stockwet
I found this a viable, safe method. The only problem is that I, in good conscience, could not ask for an interest free account on religious grounds. It goes against my integrity. Alas.

stockwet
  • #25
  • Aug 25, 2006 10:25pm Aug 25, 2006 10:25pm
Quoting ycomp
I don't think it matters... for instance neuimex offers what it advertises as "islamic swap free" accounts but I think they do that just for advertising... you have to pay a commission to trade swap free accounts with them.

I think it is 5$ per lot, I can't remember.
FXCM charges $3. I factored that into my spreadsheet, too. Actually, they really suck because they charge you $3 per lot and $1 per mini lot.
  • #26
  • Aug 26, 2006 1:32am Aug 26, 2006 1:32am
Stock,

After getting a little excited about this, I decided to think a little harder on what you said.

Each lot of the GBP currently brings in a little less than $20 a day. Over a 30 day period, that's about $600 a lot per period.

Now, let's say the spread of the GBP/JPY is 10 pips. The cost per pip is around $9 at the moment. Well, per lot traded, that's $180 in spreads (1 long and 1 short) that you are going to have to pay for every trade. If you look at the volatility in the last 30 days, we saw the movement of the GBP/JPY around 1200 pips and over 6 months the movement has whipsawed (but generally moved upwards) by 2000 pips. If what you're saying is true, that we would need to rebalance (close trades) every 350 pips, just in the last month we would have to have rebalanced AT LEAST twice. Now, if you do the calculations, $180 in spreads per lot * 5 lots = $900 and $600 a month in interest * 5 lots = $3000 which gives you a gain of $2100. Subtract $5 a lot * 5 lots * 2 trades for the interest free trades and you get a total of $2050...assuming we rebalance twice a month. This is definitely acceptible. I think where the risk comes in is when you have to rebalance 2 more times. Obviously, $1850 in spreads and trade commissions leaves you with $200 for the month.

Let's look at it this way...there are 52 weeks in a year. * $925 rebalance fee (spread + trading commissions) = $48,100 in fees a year. Assuming interest is applied 365 days in a year * $20 a day per lot * 5 lots = $36,500...this actually produces a loss of $11,600 a year.

Now, if you can rebalance only 26 times in a year, you can produce a gain of $12,450 a year or 31.1%. If you can rebalance only once a month, you get $25,400 a year or 63.5%.

The problem is, if volatility over the course of 6 months is 2000 pips, that means every month, volatility is at least 330 pips a month which means the need to rebalance occurs more than once a month, and that assumes uniform directional movement. If there are any significant pull-backs that occur immediately following a rebalance, it's entirely possible that you will end up having to increase the number of rebalances in a year...thus reducing your overall gain.

I'd like to hear your thoughts...I still think this is viable, but I think we need to figure out how to reduce the trading costs (by reducing the number of times we have to rebalance). If we cut in half, the number of lots we're buying, we can basically cut the numbers in half, BUT we also probably only have to rebalance to less than 12 times a year. So...let's see where that puts us:

(2.5 lots (25 minis) * 20 pips * $9 a pip) + (2.5 lots * $5 a lot) = $462.50 for rebalance fees * 12 rebalances = $5500 a year

(2.5 lots * $20 interest * 365 days) = $18,250 = $12,750 a year or 31.8%...that's still a decent return...and in reality, if the system could withstand 660 pip movements, there is a GOOD chance you can rebalance less than 12 times a year. In all honesty, you should let the rebalancing take care of the "gain" with a minimum of 1 rebalance a year (for new money). If you can reduce the cost of rebalancing, then you have a good system, but if you have to rebalance every month, you're going to kill the potential.

The problem here is...how long are these types of accounts going to be available? 31.8% is an good return for long-term investments...but it would still take 18 years to grow $40,000 into $1,000,000...Again, that's pretty good considering that this is about as safe as you can get in forex.
  • #27
  • Aug 26, 2006 6:21am Aug 26, 2006 6:21am
Here's my spreadsheet. Feel free to double check the forumulas and let me know if anything is amiss.

stockwet
Attached File(s)
File Type: zip swap_strategy.zip   4 KB | 381 downloads
  • #28
  • Aug 26, 2006 9:09am Aug 26, 2006 9:09am
Quoting mrmikal
Stock,

After getting a little excited about this, I decided to think a little harder on what you said.

Each lot of the GBP currently brings in a little less than $20 a day. Over a 30 day period, that's about $600 a lot per period.

Now, let's say the spread of the GBP/JPY is 10 pips. The cost per pip is around $9 at the moment. Well, per lot traded, that's $180 in spreads (1 long and 1 short) that you are going to have to pay for every trade. If you look at the volatility in the last 30 days, we saw the movement of the GBP/JPY around 1200 pips and over 6 months the movement has whipsawed (but generally moved upwards) by 2000 pips. If what you're saying is true, that we would need to rebalance (close trades) every 350 pips, just in the last month we would have to have rebalanced AT LEAST twice. Now, if you do the calculations, $180 in spreads per lot * 5 lots = $900 and $600 a month in interest * 5 lots = $3000 which gives you a gain of $2100. Subtract $5 a lot * 5 lots * 2 trades for the interest free trades and you get a total of $2050...assuming we rebalance twice a month. This is definitely acceptible. I think where the risk comes in is when you have to rebalance 2 more times. Obviously, $1850 in spreads and trade commissions leaves you with $200 for the month.

Let's look at it this way...there are 52 weeks in a year. * $925 rebalance fee (spread + trading commissions) = $48,100 in fees a year. Assuming interest is applied 365 days in a year * $20 a day per lot * 5 lots = $36,500...this actually produces a loss of $11,600 a year.

Now, if you can rebalance only 26 times in a year, you can produce a gain of $12,450 a year or 31.1%. If you can rebalance only once a month, you get $25,400 a year or 63.5%.

The problem is, if volatility over the course of 6 months is 2000 pips, that means every month, volatility is at least 330 pips a month which means the need to rebalance occurs more than once a month, and that assumes uniform directional movement. If there are any significant pull-backs that occur immediately following a rebalance, it's entirely possible that you will end up having to increase the number of rebalances in a year...thus reducing your overall gain.

I'd like to hear your thoughts...I still think this is viable, but I think we need to figure out how to reduce the trading costs (by reducing the number of times we have to rebalance). If we cut in half, the number of lots we're buying, we can basically cut the numbers in half, BUT we also probably only have to rebalance to less than 12 times a year. So...let's see where that puts us:

(2.5 lots (25 minis) * 20 pips * $9 a pip) + (2.5 lots * $5 a lot) = $462.50 for rebalance fees * 12 rebalances = $5500 a year

(2.5 lots * $20 interest * 365 days) = $18,250 = $12,750 a year or 31.8%...that's still a decent return...and in reality, if the system could withstand 660 pip movements, there is a GOOD chance you can rebalance less than 12 times a year. In all honesty, you should let the rebalancing take care of the "gain" with a minimum of 1 rebalance a year (for new money). If you can reduce the cost of rebalancing, then you have a good system, but if you have to rebalance every month, you're going to kill the potential.

The problem here is...how long are these types of accounts going to be available? 31.8% is an good return for long-term investments...but it would still take 18 years to grow $40,000 into $1,000,000...Again, that's pretty good considering that this is about as safe as you can get in forex.
Wow, Mrmikal you sure do know your stuff!
PipMasterFlex
  • #29
  • Aug 26, 2006 9:14am Aug 26, 2006 9:14am
Quoting stockwet
Here's my spreadsheet. Feel free to double check the forumulas and let me know if anything is amiss.

stockwet
Stock...your trade cost row...you didn't multiply by the number of lots. That's the big error...if you multiply that row times the number of lots it increases the trade cost 5 times. Once you do that, our numbers pretty much agree.
  • #30
  • Aug 26, 2006 9:22am Aug 26, 2006 9:22am
Quoting mrmikal
Stock...your trade cost row...you didn't multiply by the number of lots. That's the big error...if you multiply that row times the number of lots it increases the trade cost 5 times. Once you do that, our numbers pretty much agree.
No, I think I did.

D22*c7*d7*2

d22=trades per month (4)
c7=pip cost (8.52)
d7=spread
2 = do this on both accounts

Hmm. OK. I think you're right.
  • #31
  • Aug 26, 2006 9:25am Aug 26, 2006 9:25am
Quoting mrmikal
Stock...your trade cost row...you didn't multiply by the number of lots. That's the big error...if you multiply that row times the number of lots it increases the trade cost 5 times. Once you do that, our numbers pretty much agree.
See, now. Why did you have to go and shatter my excitement?
  • #32
  • Aug 26, 2006 9:30am Aug 26, 2006 9:30am
Well, if it sounds too good to be true ...
  • #33
  • Aug 26, 2006 9:54am Aug 26, 2006 9:54am
Quoting stockwet
See, now. Why did you have to go and shatter my excitement?
Trust me, I played an entire round of midnight golf with a woody about this idea.

Like I said, Stock, I think if we cut the number of lots in 1/2 we can reduce the number of rebalances requried. If we cut those costs, this would be a SAFE 20%+ APR.

The other thing we can look at is trading something like the USD/JPY, or the AUD/JPY. They all have lower transaction costs, but the interest is also a little less. The USD/JPY might produce better results...AND I think it's a little less volatile. Lower fees per trade and lower volatility = less rebalance fees year over year. Even though the interest may be lower, it may be in our interest to research that? I dont' have time today to fudge with your numbers, but man, I'd appreciate it if you could do your calculator do hicky and see what we get.

My conservative calcs are:

Interest is $13 a day or so. That's $23,725 a year

The fees per trade on 5 lots would be $295 per rebalance * 52 rebalances = $15,340.

Interest gained is $8,385 or 20.9% a year...and that's assuming 52 rebalances a year. If you reduce the number of rebalances to 26 then you're looking at a 40% a year gain. Based on the volatility I think the number will be somewhere in that range (between 26 and 52).

Please double check my math. Thanks.
  • #34
  • Aug 26, 2006 10:18am Aug 26, 2006 10:18am
Here's the gross monthly return for each, assuming 5 lots on each side and refreshing 2 times per month:

GBPJPY: 2.66%
GBPCHF: 1.59%
USDJPY: 3.15%

Even if you refreshed JPY 5 times a month, you're looking at a monthly return of 2.13%.

The other thing to look at is increasing the leverage - finding one or both accounts with 1:400 leverage. That would enable wider swings. But, USDJPY is still appealing.
  • #35
  • Aug 26, 2006 10:20am Aug 26, 2006 10:20am
You know, there has to be a way to do this so that you are hardly ever refreshing. I bet you could set up a process where you have funds in reserve. When you get close to margin on one account, you wire the funds to one and take funds out of the other. You'd get hit with a $15 wire charge each time, but, that's still cheaper than the cost of closing and opening orders.
  • #36
  • Aug 26, 2006 11:27am Aug 26, 2006 11:27am
Quoting stockwet
You know, there has to be a way to do this so that you are hardly ever refreshing. I bet you could set up a process where you have funds in reserve. When you get close to margin on one account, you wire the funds to one and take funds out of the other. You'd get hit with a $15 wire charge each time, but, that's still cheaper than the cost of closing and opening orders.
you could keep up by using a daily ATR. if it takes 3 weeks to mail in checks (so you dont have to wire) then you could do ATR * 4 weeks. but really if you keep 2000pips worth of equitity avaible you stil end up with 35% a year and you only need 1 account transfer per month and you dont have to open/close orders.
  • #37
  • Aug 26, 2006 11:34am Aug 26, 2006 11:34am
Quoting stockwet
You know, there has to be a way to do this so that you are hardly ever refreshing. I bet you could set up a process where you have funds in reserve. When you get close to margin on one account, you wire the funds to one and take funds out of the other. You'd get hit with a $15 wire charge each time, but, that's still cheaper than the cost of closing and opening orders.
Keep thinking on that track man...we only want to rebalance 1 or 2 times a year...But we want to make sure and keep the lot sizes the same.

I can't really think of a good method other than having more money on the sidelines that merely gets shuffled in and out of accounts...so in other words, if you had, say $60,000 to use...you could put $20,000 in 1 account, $20,000 in another account and then move $20,000 in and out of accounts as the margin call nears...thus making an unbalanced fund. At 5 lots, each pip is worth $45 or so, so you could withstand another 400 pips until you had to rebalance. That would drop your percentage gain a little bit, but you'd be withstanding larger moves.

The only other thing I can think of is only liquidating one side to realize the profit and then immediately buying back. Even though the price won't be the same one one side of the account, the hedge order will still be (as 1 account rises 1 pip, the other falls 1 pip). SO...I dunno...you may only have to liquidate 1 side of the position during a rebalance.

even if we spent $100 in wire fees, that's a cheaper rebalancing method.

Keep thinking stock...between you and me, I'm sure we can come up with a better way of doing this...
  • #38
  • Aug 26, 2006 12:21pm Aug 26, 2006 12:21pm
Quoting witchazel
you could keep up by using a daily ATR. if it takes 3 weeks to mail in checks (so you dont have to wire) then you could do ATR * 4 weeks. but really if you keep 2000pips worth of equitity avaible you stil end up with 35% a year and you only need 1 account transfer per month and you dont have to open/close orders.
That's an interesting way to keep track of volatility. It would give a good indication for timing account transfers.

Keeping 2000 pips available is a nice idea. It sure does cut into the amount of cash you tie up, though. I'm having trouble believing that one couldn't get more than 35% per year trading their own money with normal strategies. Of course, you could get it to a point where it's risk free. So, that's a bonus.
  • #39
  • Aug 26, 2006 4:45pm Aug 26, 2006 4:45pm
Quoting witchazel
you could keep up by using a daily ATR. if it takes 3 weeks to mail in checks (so you dont have to wire) then you could do ATR * 4 weeks. but really if you keep 2000pips worth of equitity avaible you stil end up with 35% a year and you only need 1 account transfer per month and you dont have to open/close orders.
2000 pips * 5 lots * $9 a pip = $90000 + equity required @ 400:1 = $92,500 in one account. So that's $185,000 for 2 accounts.

Now, let's say the pip cost is back to $925 a transfer and we rebalance only once a year.

Interest is $20 a day * 5 lots * 365 days = $36,500

So...the total gain rebalancing only once a year is $35,575. Thus, we're looking at a 19.2% gain.

Now before any of you freak out about having to have $185,000...just look at the return and divide it by how much you do have and figure it out.

Now, let's just say we decide to keep $90000 in equity somewhere (between the 2 accounts)...the total equity requirement for 5 lots would be $95,000.

Now, excluding wire fees, the amount gained would be the same, but now we're looking at 37.4%. So all we need to do ever is shift money from one account to another during the year. I suspect we would only need to do this a few times a year. Let's just say we have $1000 in wiring fees, we're looking at 36.4% a year. That's not too bad.

Here's the big risk, folks...rate increases. We are assuming that the BOE and the BOJ will either raise rates in-synch with each other. Unfortunately, it looks like the BOJ is taking a more aggressive stance than the BOE (although we've been surprised before). If the BOJ continues to raise rates and our target rates do not increase at the same (or faster), our interest potential shrinks quickly. As easy as this sounds, It IS possible that the system will not be viable for lengthy periods of time (like if the BOJ ever raises rates comparable to the US or the BOE)...again, it's hard to think of a scenario that would cause that to happen, but this isn't a sit on your hands kind of system...we have to be extremely aware of how the rate changes impacts our ability to earn interest or we'll be sitting on a lame 4% a year return.

Keep thinking guys...everyone is coming up with really good suggestions.
  • #40
  • Aug 26, 2006 6:37pm Aug 26, 2006 6:37pm
you are simply talking about having 1000pips in reserve on the outside. thats works fine for me too. the balance is to find the minimum you need in reserve so you dont ever have to re close/open the position.

if you were truely brave you could go for 500 pips and make 70%. its a balancing act between reentry and APR

If you plan ahead you dont need to puy wire fees either, you can use the post office.

i prefer 2000pips as this isnt my main trading. i only have 10% of my account actively trading at any time, these 2000pips are the other 90%. its nice to have a 30%+ APR on stuff you arent even actively trading with
0 traders viewing now