Search results

The captured tweets are at these links and can be accessed from outside of the forum (unlike attachments). But no one can edit. https://www.s4me.info/docs/SharpeTweets1.pdf https://www.s4me.info/docs/SharpeTweets2.pdf https://www.s4me.info/docs/SharpeTweets3.pdf...

Another batch of screen captures (thanks to @Keela Too ) https://www.s4me.info/docs/SharpeTweets4.pdf https://www.s4me.info/docs/SharpeTweets5.pdf https://www.s4me.info/docs/SharpeTweets6.pdf https://www.s4me.info/docs/SharpeTweets7.pdf

@Keela Too put together some documents of the tweets they can be got here https://www.s4me.info/docs/SharpeTweets1.pdf https://www.s4me.info/docs/SharpeTweets2.pdf https://www.s4me.info/docs/SharpeTweets3.pdf

From the TSC committee meeting #8 (page 10) https://www.dropbox.com/s/1b86k0pvnymko16/TSC.zip?dl=0&file_subpath=%2F20090429+TSC.pdf "The need to carefully select the person who writes the editorial on the main paper was also discussed."

No the forum just has the URL to the tweet. Linking to the tweet causes it to be embedded in the web page by your web browser getting the content from twitter.

I think that when he publicly jumps ship we know we have won. He seems to do a fair bit of behind the scenes lobbying and I suspect that is still going on. But I think he will be concerned for how he is viewed in a historical context.

Would a positive outcome give support for their model? If there were other potential models that were equally plausible then it would only provide support as one of a set of possible models. So to design a good experiment you need to think of other plausible explanations and design a hypothesis...

g I think they refer to a model in their CBT and GET manuals. But I don't think I would call it a model as its a few boxes with lines between them. That is I don't think there is enough to tie down what they are saying.

I don't think Sharpe understands what is wrong. Maybe he really does believe that the paper explains everything and should convince people. But it seems that when outsiders look they pick holes and don't like the methodology.

I've corrected the version attached to the start of this thread and also the one in the https://www.s4me.info/docs/PaceBriefing3.pdf URL.

I think I tend to agree. Charities could raise a lot for research and also an effective way to increase government research funding it to seed some initial research that can be turned into bigger projects that can apply for government funding. One of the issues raised by government is lack of...

As with any other research that tries to use the CFQ to measure effect I would say it has no value. The CFQ is simply not capable of measuring anything in a coherent way.

I thought that there was an interesting philosophical point being made in terms of whether one can do evil without being evil. Maybe evil is too strong a word but it brings out a point around whether an act that is wrong need be done with bad intention and I think this does relate to this debate.

They are still hiding data. [Added] Its also if the senior university staff who don't control standards fail and get away with it there is no pressure for others to enforce standards. They get paid a high wage to perform such tasks.

I wasn't aware it came directly from the Vice Chancellor. That puts him in a difficult position in the long term because what he has done there fails basic ethical tests (certainly any headline test). It is something that eventually could stick to him and destroy his reputation for a long time...

To me its less the authors who should face consequences and more those who failed to apply a level of governance. So for example the head of QMUL who failed to ask questions and spend a lot of money trying to cover up the results. Or the Vice Chancellor of Bristol university who has failed to...

So what is a pragmatic trial?

So the impression I get from your paper (which could be wrong I only read it quickly) is that you gave a supplement to a number of patients with a disease that is varies over time and then sorted the patients into those who improved in a given time period and those who didn't. It seems hard to...

I think its something to do with the maximum likelihood optimization over what are assumed to be probabilities using the cost function: J(θ)=−∑i(y(i)log(hθ(x(i)))+(1−y(i))log(1−hθ(x(i)))) With y(i) being the target ith training sample and x(i) being the input for the ith training example and hθ...

So I would come back to my question of why the output of a logistic regression would be a probability? Does it meet the basic axioms. My assumption is not but i've not looked into it. I did read about logistic regression a few years ago and it seemed to be similar to a single layered neural...