tag:blogger.com,1999:blog-3932338495391547190.comments2024-01-06T10:17:50.985-08:00factbasedKarsten W.http://www.blogger.com/profile/05710598626358932856noreply@blogger.comBlogger30125tag:blogger.com,1999:blog-3932338495391547190.post-87469179638097662912024-01-06T10:17:50.985-08:002024-01-06T10:17:50.985-08:00This was lovelly to readThis was lovelly to readCupcakes Are So Coolhttps://cupcakesaresocool.tumblr.com/noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-92200531887685795972023-09-02T16:49:35.027-07:002023-09-02T16:49:35.027-07:00Good reeading your postGood reeading your postsakuraso blog animehttps://yukinoyukinoshitaa.blogspot.com/noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-24719743650918348212023-02-28T08:09:30.607-08:002023-02-28T08:09:30.607-08:00Nice post. Yeah, ChatGPT can do amazing things, bu...Nice post. Yeah, ChatGPT can do amazing things, but not reliably yet. Will be very interesting to see how it develops. I'm a bit on the skeptical side, as it does not seem to have proper language understanding.StatistikinDDhttps://www.youtube.com/@StatistikinDD/videosnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-36992004493587103102023-02-10T02:22:36.036-08:002023-02-10T02:22:36.036-08:00When the AI suggests wrong code again and again, t...When the AI suggests wrong code again and again, that is a No-go for me. Hopefully, some competition will rectify that.Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-22004318780201721922023-02-08T22:53:13.001-08:002023-02-08T22:53:13.001-08:00ChatGPT is good at regular expression related thin...ChatGPT is good at regular expression related things and some basic R questions. However, it fails miserably in most R flows that cannot be done in one line. I'm quite fed up with "Yes, you are correct... Here is a corrected code..." ... which still has fundamental errors. Not the quickest way to solve those R problems. Less R code available for training it?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-27246931101006186812022-10-31T08:42:43.238-07:002022-10-31T08:42:43.238-07:00You might want to include your mastodon address. F...You might want to include your mastodon address. For example, mine is @markniemannross@mastodon.social<br /><br />MNRmnrhttps://www.blogger.com/profile/17564728029005756503noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-2617512362699634972022-08-28T15:21:46.428-07:002022-08-28T15:21:46.428-07:00Hi great reading yourr postHi great reading yourr postFencing Contractors Citrus Heightshttps://www.fence-contractors.com/us/fence-builders-california/fencing-contractors-citrus-heights.shtmlnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-50625691362824544502022-01-15T03:29:35.950-08:002022-01-15T03:29:35.950-08:00Good reading your ppostGood reading your ppostTyreesehttps://www.tyreesenelson.com/noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-68079129519765122992021-07-06T01:15:53.267-07:002021-07-06T01:15:53.267-07:00This comment has been removed by a blog administrator.nencyhttps://www.blogger.com/profile/06465734486094976925noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-62087915234639877852020-12-11T05:03:52.973-08:002020-12-11T05:03:52.973-08:00This is great! I've been looking to do somethi...This is great! I've been looking to do something like this, so it's cool to see there's others before me who have made it happen... Even if it was almost a decade ago 😅<br /><br />Couple questions:<br /><br />Are you still using this system? If not, how long did you run with it and what did you end up switching to? Did you use Twitter the whole time or switch to something else for tracking cash?<br /><br />I've been throwing around the idea of using a Termux script/form for capturing expenses, and having them auto-log in my system. I like that because, in theory, I could have it commited to a git repo that I then pull from when processing later on my computer. Then I can import it into R and we're away to the races. The effort on my side is <i>1) tap script shortcut, 2) type expense and detail</i>, which seems to be easy enough that I'll actually do it.<br /><br />Do you have any more wisdom that you've gleaned since you did this in 2011? Be great to hear an update.<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-36644316777854599532019-12-11T04:25:01.076-08:002019-12-11T04:25:01.076-08:00No, it's base::plotNo, it's base::plotKarsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-57938386284950150202019-12-11T04:20:33.953-08:002019-12-11T04:20:33.953-08:00Hi, do you have a custom "plot" method? ...Hi, do you have a custom "plot" method? Or how does it accept dat=1 and dc_dat=dc_dat? Thank you!Felix Dietrichhttps://www.blogger.com/profile/09511327388326985836noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-21934057625500122832019-03-02T05:47:40.170-08:002019-03-02T05:47:40.170-08:00try this, loop can be moved to C/C++ for even fast...try this, loop can be moved to C/C++ for even faster processing<br />```r<br />library(data.table)<br /># priceframes - directional changes<br />get_priceframes = function(x, threshold) {<br /> stopifnot(is.double(x), is.double(threshold))<br /> nx = length(x)<br /> if (nx < 2L) stop("provided input is length ", nx)<br /> imx = imn = 1L<br /> trend = open = confirm = rep(NA_integer_, nx)<br /> last_trend = 0L<br /> for (i in 2:nx) {<br /> if (last_trend < 1L && x[i] >= x[imn] * (1.0 + threshold)) {<br /> trend[i] = last_trend = 1L<br /> open[i] = imn<br /> confirm[i] = i<br /> imx = i<br /> } else if (last_trend > -1L && x[i] <= x[imx] * (1.0 - threshold)) {<br /> trend[i] = last_trend = -1L<br /> open[i] = imx<br /> confirm[i] = i<br /> imn = i<br /> }<br /> if (x[i] > x[imx]) imx = i<br /> if (x[i] < x[imn]) imn = i<br /> }<br /> d = as.data.table(list(<br /> trend=trend, i_open=open, i_confirm=confirm<br /> ))[!is.na(trend)][, "i_close" := shift(i_open, -1L)][]<br /> icols = c("open","confirm","close")<br /> d[, paste("price", icols, sep="_") := lapply(.SD, function(i) x[i]), .SDcols=paste("i", icols, sep="_")][]<br /> d<br />}<br />```Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-51937020577784533382019-01-27T14:16:51.376-08:002019-01-27T14:16:51.376-08:00Added a link to tick data, thanks for the hint to ...Added a link to tick data, thanks for the hint to bitcoincharts.com. The code is now fixed, but seems to be too slow for my taste.Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-44396816645250238872019-01-25T07:57:55.825-08:002019-01-25T07:57:55.825-08:00Yep, that seems to be an error. Will fix it on wee...Yep, that seems to be an error. Will fix it on weekend.Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-23116351918779306792019-01-25T07:33:31.502-08:002019-01-25T07:33:31.502-08:00Thanks. Good resources of tick trades easily avail...Thanks. Good resources of tick trades easily available can be found at https://api.bitcoincharts.com/v1/csv/<br />Regarding your code, the else branch in while loop below is not always what the comment suggests. `mode` value is initialized as "uoi/dei" or "doi/uei". Is it OK that `else` kicks in for those values then?<br /><br />if(mode=="dcc_up") {<br /> ...<br />} else { # mode=="dcc_down"<br /> ...<br />}<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-30312523967175977152019-01-25T07:12:46.734-08:002019-01-25T07:12:46.734-08:00Hi, thanks for reading & commenting. The code ...Hi, thanks for reading & commenting. The code should work for tick trades. I used OHLC data because I do not have other (yet). I plan to research data this weekend and post it if public available.Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-51053957465959421252019-01-25T05:18:40.785-08:002019-01-25T05:18:40.785-08:00Interesting post. Your code seems to expect OHLC t...Interesting post. Your code seems to expect OHLC trades data, which already contradicts the idea of forming tick trades into price frames rather than time frames. Could you comment on that? should the code work for tick trades? Example input data have not been provided, maybe you could link a gist to that so your analysis will be reproducible. Thank you!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-33510689701758436282018-04-02T15:28:54.597-07:002018-04-02T15:28:54.597-07:00Hi,
This is super useful!
How can I get the full ...Hi,<br />This is super useful! <br />How can I get the full code?<br />Thanks! NatanMhttps://www.blogger.com/profile/12569565256140703461noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-35288715143943204502016-11-23T21:11:55.123-08:002016-11-23T21:11:55.123-08:00Hi Mr. Weinert!!
Your package, ida, is crantasic!...Hi Mr. Weinert!!<br /><br />Your package, ida, is crantasic!!<br /><br />I tried to install ida, but it doesn't work.<br /><br />Would you please let me know how to install and apply the ida package.<br /><br />Please mail me jjang [at] nfri.krHansoo CHANGnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-7163427107367809292015-07-19T04:56:33.766-07:002015-07-19T04:56:33.766-07:00I think, data scientist is someone who can find th...I think, data scientist is someone who can find the secret in data and predict the future, they are somehow like the fortune teller, but based on the science and data.Connyhttps://www.blogger.com/profile/08456765212209184596noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-28111299032036052222014-06-17T23:33:19.961-07:002014-06-17T23:33:19.961-07:00thanks for asking. I am planning a new release of ...thanks for asking. I am planning a new release of the package in June. there will be an area and volume unit set then.Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-50446597268651363042014-06-17T15:22:13.136-07:002014-06-17T15:22:13.136-07:00This is very helpful! Do you have any plans to in...This is very helpful! Do you have any plans to incorporate volume units to your uconvlist? Or do you know any functions that do this?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-72204209453328049602013-02-20T12:17:34.155-08:002013-02-20T12:17:34.155-08:00Thanks, dinre. I find the negative sign in the PV ...Thanks, dinre. I find the negative sign in the PV and PMT functions hard to track. It makes sense when I think about what you wrote, I hope it stays in my brain...Karsten W.https://www.blogger.com/profile/05710598626358932856noreply@blogger.comtag:blogger.com,1999:blog-3932338495391547190.post-23759869277667542662013-02-19T06:24:26.047-08:002013-02-19T06:24:26.047-08:00Perhaps the easiest way to remember the meaning of...Perhaps the easiest way to remember the meaning of the sign of the internal rate of return is to remember the basic, numerical form of the solution:<br /><br />Sum { Cashflow(Period) / (1 + IRR) ^ Period } <br />=<br />Initial investment<br /><br />We are trying to find the one rate of return value that will make all of the periods' cash flows add up to the initial investment.<br /><br />Since we are dividing by the rate of return, positive IRR is another way of saying that our cash flows are bigger than the initial investment and need to be adjusted down to solve the equation. Negative IRR is saying that the cash flows are smaller than the initial investment and need to be inflated.<br /><br /><b>The overly-simple way of thinking about it</b>:<br />IRR is (loosely) the change in your account balance after you run a project. Negative means you will lose money, and positive means you will gain money.<br /><br />I've never seen a negative IRR in real life, because those projects don't make it past initial project screenings. Usually, IRR is used as one measure to see if the opportunity cost of doing the project is outweighed by the return, so IRR is somewhat meaningless without understanding the opportunity cost. Often, the opportunity cost is your company's projected borrowing rate, and when borrowing costs more than your project's return, you don't do the project.<br /><br />Hope that helps.Anonymousnoreply@blogger.com